okay so today we're going to talk about psychological research and how psychology fits into being a science if you remember last week we introduced the definition of psychology which defines it as a science um that studies mental processes and human behavior and so during this chapter I'm going to introduce to you guys why it's so important that psychology is scientifically based science is built on a set of core beliefs about the world and there are two key beliefs the first one is that the Universe operates according to certain natural laws scientists believe that things that happen in and around us happen in some kind of orderly fashion and that can be described using laws or rules for example the natural law of cause and effect suggests that when something is set in motion it has an effect on other things psychologist look for the laws that describe mental processes and behavior number two such laws are discoverable and testable by carefully observing what happens in the natural world we can figure out the laws governing those events and in turn we can use those laws to make predictions about what might happen in the future and we can then experiment to see whether those predictions will come true as a natural science psychology operates according to these two core beliefs psychology also shares with other Sciences a similar logical approach to discovering and testing laws about how things happen and that is the scientific method just as a refresher the steps within the scientific method are observe hypothesize test conclude and evaluate and I'll give you an example of each of these on the next slides the first step in the scientific method is to observe and this step is typically triggered by a curiosity curiosity is about the researchers motivation for the entire process of the scientific method they're curious about something and they want to find out the answer to their curiosity for this example I'm going to use honeybees let's say I'm hanging out in my backyard and I'm watching the honeybees in my backyard go from plant to plant and I notice that honeybees seem to spend more time in squash flowers than they do on lavender flowers and so I want to find out why and so my curiosity about why bees spend more time on a squash flower than a lavender flower is the motivation behind this research the variables are the things that I'm measuring and looking closely at within my research so the variables in this research would be the time spent on a squash flower or a lavender flower by the bee and then the type of flower that they're spending time on a squash flower or a lavender flower and then the theory is about trying to understand why the observed thing happened so my theory and I really don't know anything about this but I'm going to assume or theorize that bees tend to spend a longer time in squash flowers because squash flowers are larger so maybe they have more pollen and it's also just a nice little protective flower to hang out in or sleep in or take a rest in compared to a lavender flower that's very exposed and there it looks like there may be less pollen so that is my theory so the next step is to create a hypothesis so the most important thing about creating a hypothesis is that it needs to be able to be tested untestable predictions cannot be investigated you have to be able to measure what your witnessing in order to see if any change has occurred and you have to be able to measure it so that you can replicate the research in the future that's a really important step to the scientific method so the way that the hypothesis that I'm going to create about my honey bees is that honey bees will spend twice as much time in in squash flowers than they do on lavender flowers that's it that's my hypothesis super simple and I can measure it how can I measure it well I'm using time and time you can measure so I'll sit there and I'll watch the honeybees and I'll record how long honey bees spend on squash flowers and how long they spend on lavender flowers and remember a hypothesis without a theory does not help build a larger picture of the world so my theory about why honeybees seem to spend or may spend more time in squash flowers is connected to this hypothesis because my hypothesis will prove if they really do in fact spend more time on squash flowers than on lavender flowers will it tell me exactly why they spend more time in squash flowers than lavender flowers no but it's a part of the big picture which is that they do hopefully we'll see they may spend more time on a squash flower than a lavender flower and the next step is to test my hypothesis so now that I have these variables that I'm going to measure which is going to help me figure out if my hypothesis is right or not I'm going to actually go out in my backyard and test these questions that I have so the way that I'm going to do that is I'm going to have my timer or my phone ready and I'm going to watch bees on my squash flowers and I'm going to see and record on my phone and then write down on a piece of paper how long a bee spins on a squash flower on average right so I would just write down the amount of time the bees spends on the squash flower and maybe I'll watch this for an hour I'm not really going to do this because it sounds extremely boring but this is just for an example um and then right so that's measuring how often or how long bees spend on a squash flower then I'll go I'll move closer to the lavender Bush and I will measure and watch and record to see how long bees spend on a lavender flower so that is how I'm going to test my hypothesis it was very easy for me to collect this data because I'm using time and time is measurable not all variables are this easy to measure let's say you're doing a research and you want to measure happiness maybe you want people to um try this anti-depression pill and you want to see how it affects people's happiness how would you measure happiness well luckily people have created um surveys that you can you can give to people to measure their happiness you can also measure happiness other ways like how many times someone Smiles how often someone says hello to someone um so there's ways to measure happiness but it's more complex than just something like measuring time so we're going to learn a lot more about how to make things measurable make variables measurable um later on in this chapter but for my own hypothesis and test that I'm conducting in my backyard with my honeybees it's a very easy test because I'm just looking at time so after I collect my data from watching bees for two hours and and measuring and timing how long they spend on squash flowers and how long they they spend on lavender flowers and analyzing that data right I'm going to probably add up all of my all of my times that be spent on squash flowers and then average them to see how long on average a bee spins on a squash flower and then I'm going to do the same thing with all of the times recorded for bees that the bees spent on the lavender flowers and find the a average time be spent on a lavender flower um and then I'm going to think about my data and analyze it and see if it aligns with my hypothesis does do my results support my hypothesis how do the results alter my original Theory can we trust my conclusions can these findings be generalized to other people in other settings doing the same research so these are all things I'm going to want to talk about in my conclusion did my results prove that honeybees spend twice as long on at least twice as long on a squash flower as they do on a lavender flower does that support my theory that bees hang out in squash theories s squash flowers longer than lavender flowers and if someone were to redo my research some somewhere else not in my backyard maybe in their backyard would they find the same results so can can this information be generalized to all honeybees at least in my neighborhood what about in my city what about in my state what about in for honeybees worldwide so all of these things are really important in um in your conclusion to think about the significance of the results and how they fit into the world and what they're actually saying so the last step of the scientific method is to evaluate evaluate your study how did you do did you do did you do things right so one thing you can do is you can um adjust your theory based on your results right so my theory was that on average honeybees spend twice as long on a squashed flower than they do on a lavender flower and I don't know what the result is because I haven't conducted the stud St but based on my findings I will know whether I need to adjust my theory or not what if I I I look at my data I noticed oh wow honeybees spend five times as long on a squash flower than a lavender flower so I would have to change my theory and then what if I found that oh wow like actually the average honeybee spends less time on a squash flower than on a lavender flower right that's going to change my theory dramatically that's going to completely prove my hypothesis wrong so you want to look at that you're also going to want to get peer reviews so someone in the field um who can review review your research and make sure that you didn't make any mistakes that you weren't being biased that your design was a good research design you followed the scientific method um that's really important it's kind of like when you write a paper and and you can read it over and over and then you let someone else read it and they find like a spelling error that you just didn't see it's just getting a second set of eyes to make sure that you you all the boxes are checked and you did everything the right way um you'll also want to do a meta analysis um which is to compare your data um with other statistical analyses of similar several similar research experiments but not yours right so other other research that also looked at the time honeybees spend on squash flowers compared to lavender flowers I don't even know if I would ever find a research like that to be honest with you but these are things you want to think about when you're even thinking about conducting a research like can I will I be able to do a meta analysis on this like are there other experiments out there measuring the same thing looking at the same thing that I can compare my data with and and this the per the purpose of doing a meta analysis is to validate your your your data because if you find other research that has the same outcome as you it helps detect statistical significance within your data it's showing that your data is is correct your study was done right and what you're finding is the truth you didn't screw it up it's not biased what your data is showing is the true reality of honeybees and how how much time they spend on squash versus lavender flowers so how do psychologists conduct research well they go through the scientific method which we just covered um but we're going to go into more detail about this because um it's really important that you know certain details about the scientific method and how it is used in Psychology because psychology is different it's different than just um a regular science so we're taking something that's super complex and very difficult to Define and measure things like human thoughts and behaviors and and thought processes and emotions and feelings those things are more difficult to research and measure than like the weather right so that's why it's important to really look further into how this is done um and so we talked a little bit about variables and how your research needs to have variables and that's what you're looking at to see um to measure to see if to see what's going on so my variables I'm going to stick with my honeybee example because um because we're we're we're all the way in with it I'm just going to keep going with it um and my variables in the honeybee example were the type of flower that the honeybees were visiting so either a squash flower or a lavender flower and then the other variable was how much time was spent on each of the flowers so variables can be organized into two different categories we have an independent variable which is the variable that um usually is either presented or taken away or increased or decreased um but it's independent meaning it's not going to be changed by anything it's it's just there and it's present and it's usually the thing influencing the other variable which is called the dependent variable and that makes sense right so the dependent variable is dependent on the independent variable so in my example the dependent variable would be the amount of time honeybees spend on each flower and the amount of time that honeybees spend on each flower is going to depend on on which kind of flower it is which is the independent variable either a squash flower or a lavender flower so you should totally practice this on your own I give you three examples and I want you to try and figure out what the independent variable is and what the dependent variable is in each of the following experiments so the first experiment Dr Smith examines how daily exposure to a sunlamp one hour of exposure or no exposure impacts people's depression levels in the winter time so what are they measuring here and what are they using to influence what they are measuring if you guessed that the dependent variable is people's depression levels then you're right the daily exposure to a sunlamp is the independent variable why do we know this well we know that the dependent variable is usually dependent on the independent variable and so what we're looking at is to see if depression levels are affected by sunlamp exposure and so depression levels are what we me we are measuring and they're what depend on the sunlamp exposure so they're the dependent variable um and notice something interesting about sunlamp exposure in the parenthesis they explain very clearly um what sunlamp exposure means in the experiment right if they didn't have that in there would you be able to replicate this experiment we wouldn't know we wouldn't know how to replicate it we wouldn't know how much sunlamp exposure participants were getting and so what they did is they told us they said 1 hour of exposure or no exposure and this is a perfect example of how they're operationally defining the variable that is the sunlamp exposure okay so the next experiment asks does a child age four five or six influence sorry does an age of a child four five or six influence when children can learn to read and so first thing you're going to want to do is look at what are they measuring here so it seems like they're measuring when children learn to read what's affecting that well they're looking to see if age affects that so that means that the dependent variable is when children learn to read and what it depends on is their age making age of of the child the independent variable and notice again they um were very they were very clear on the age so they operationally Define the age of the child four five or six the last experiment does the amount of time reading your textbook every day once a week or none influence how you will do on the exams what are we measuring here it seems like they're measuring exam performance and what do exam performance depend on in this experiment well it depends on the time spent reading your textbook so that would make exam performance the dependent variable because it's dependent on the indep independent variable of time spent reading your textbook and notice they also um operationally defined what that means what time reading your textbook is in that experiment every day once a week or none some tip to help you guys if this is confusing for you um the first thing that you can always do when you um are presented with an experiment is to just find the two variables find the two things that are being studied so let's take the sunlamp exposure as an example so in the sunlamp experiment the two variables are very obvious it's sunlamp sun sunlamp exposure and depression levels okay those are the two variables but if you're struggling to figure out which one's independent and which one's dependent try to just try to mess around with them okay for example like I said earlier can the sunlamp exposure depend on or be affected by someone's depression no because that that doesn't even make sense can someone's depression be affected by sunlamp exposure yes that makes sense and there you go again we already talked about this but um operationally defining variables is a huge and important step in doing doing research because it allows you to measure your variables to see what changes took place or to see what took place in general and it also allows someone else to replicate your experiment exactly how you did it so an operational definition is how we the researcher decide to measure our dependent variable so my dependent variable in my honeybee experiment is how much time honeybees spent on a squash flower compared to a lavender flower um and so I measure that by using time and maybe I'll use seconds or minutes or both to operationalize is to develop a working definition of a variable that allows you to test it um there are usually hundreds of ways to measure dependent variables it just gets more difficult to oper opal operationally defin variables that are very complex like behaviors are easier to operationally Define than maybe emotions right because behaviors are observable we can see them with our eyes most of them can we see emotions we can to a certain extent using or watching body language and facial expressions but it's more complex and it's more difficult because we all know that people might be feeling emotions and not showing them so it gets a little more complicated operationally defining variables can be really difficult um but a lot of time needs to be spent on this it's a really important step when we do research we have to decide how we're going to measure the dependent variable and so I've decided that the easiest way for me to measure how long honeyb spend on flowers is by using time and by timing them in minutes and seconds so we talked in the last slide about sunlamp exposure how might you operationally define sun sunlamp exposure and depression levels from the previous example well the example already gave us a way to operationally define sunlamp exposure remember it was 1 hour or no hours or no time whatever it said but how would you how would you measure depression levels how would you do that there's tons of ways you can actually do this because people have already thought about this and people have already wanted to measure depression for other research experiments and so there are depression inventories that are highly reliable and valid which means that they are proven to accurately measure depression levels in people and it's an assessment or an an inventory is really like a survey that someone a set of questions that someone answers and it measures their depression levels um so that's something that you can use to measure depression levels so knowing what you know now about operational definitions take some time and think about how you would operationally Define the following three variables self-esteem shyness and memory loss self-esteem is much like depression um in regards to how you can measure it so there are a ton of self-esteem inventories or surveys right that you could give people that ask them a set of questions that accurately measure their self-esteem levels so that would be a good choice shyness on the other hand and is a little more complex and difficult to measure you might have to get creative in how you measure shyness think about it how would you measure shyness one way you can do this is think about a shy person how do they behave or what do they do or not do that people who aren't shy do something that I've noticed and this is because I'm a shy person shy people typically don't initiate conversations with strangers so that's one way you could measure shyness how many strangers they in they initiated conversation with per day and this is just an idea you might instead want to use eye contact because shy people tend to make less eye contact with strangers than people who are less shy or not shy so maybe how many strangers they made eye contact with each day and then memory loss is a little easier to measure because there are memory quizzes um there's memory inventories that measure how well people remember things you could also give people your own memory test by asking them or showing them a set of information or reading them a story and then asking them to recall important topics or points made in the story after some time has gone by so there's a ton of different ways you can measure memory loss I think it's always best to try and find a wellestablished and highly used inventory that it has been proven to be reliable and and valid um that's usually a very safe route to take but sometimes things like shyness they don't have inventories so you have to get creative the next really important step is you want to choose your participants and what's really going to guide this decision is who were you interested in so I'm interested in honeybees in my backyard that's the um those are the participants in my study so the population is the entire group that of in that's of interest in to the researchers so my population is honeybees now am I going to be able to collect data from every single honeybee in the world no am I going to be able to collect data on every single honeybee in my backyard no that's impossible I would never be able to do that I can't even tell the difference and I'm really hoping that I don't sit there and measure the same bumblebee or honeybee over and over because they all look the same so I'm going to pick a sample of honeybees in my backyard since I can't measure all of them and a sample is a portion of any population that is selected for the study and that will represent that population so my sample of honeybees is going to be any honeybees that I watch or or come into contact with hopefully not physical contact between between 100 p.m. and 400 p.m. that will be my sample um so it's just a it's just a portion of the population of honeybees it's not all of them and that's usually how all research goes there's no way for any research to I mean it's very difficult to to be able to have everyone in the population participate in the in the experiment and we're going to talk a little bit more about this on the next slide and then we have random selections so identifying a sample in such a way that everyone in the population of Interest has an equal chance of being involved in a study so I'm not going to sit there and cherry pick which bees right I'm not going to only only watch the bees spending a ton of time in the squash flowers it needs to be random it needs to be maybe I'll pick a one squash flower instead you know that way it's not it's fair that all honeybees that are flying around between 1: and 400 p.m. have the chance to land on that squash flower I can't just pick handpick cherry pick the bees that I'm that I'm deciding to include in my research like I'm going to pick the really big ones or I'm going to pick the ones that um are spending the most time in the squash flowers because what does that do that just screws with my data and it's not fair it's not accurate because I'm manipulating things and I'm being biased the next thing is sampling biases which I just mentioned so selecting a group that is especially likely to confirm your hypothesis this is wrong and you shouldn't do it so I should not only keep the time data from the honeybees that spent the longest amount of time on the squash flowers I should also include data from honeybees that spent a couple seconds on it you know you're not collecting data that only validates your hypothesis because that is not a true representation of what's going on in the world and the whole point of conducting research is to find truth so here is a little in-class practice to help you distinguish between a sample and a population because there seems to be some confusion over this um typically when I'm going over this in class so here are some groups of people listed and I want you to decide if it's a population or if it's a sample of the population so the first one is professional football players is this a population or is it a sample so if you guessed population you are correct the reason why it's the population is it because it includes everyone in that population there's like no one being left out professional football players is the entire population it's professional football players now let's say it said 10 professional football players from each NFL team would that be a population or a sample well it would be a sample because it's only a portion of the population next we have 10 students from Harvard University well that's an easy one it's definitely a sample why because it's not including everyone in the population if it said students from Harvard University that would be a population but it's a sample because it's only including 10 students from Harvard University the next one is albino rats obviously this is a population right and is there any way on Earth that anybody could collect data from all albino rats no that's impossible so that's why you have to select a sample so maybe would say 500 albino rats in the northern California area that would be a sample of the albino rat population next we have 100 adults over the age of 65 this is a sample because it's only pulling 100 adults from the population of people that are over the the age of 65 if it just set adults over the age of 65 that would make it a population and next we have 90 preschoolers this also is a sample from the population of preschoolers if it just said preschoolers it would be a population but because there's a number that means that they're pulling a small sample from that given population so there's two types of research um there's descriptive research and there's experimental research experimental experimental research is on its own and then there's all this other type of descriptive research that we're going to learn about and psychological research includes both kinds of approaches because descriptive methods and experimental methods each serve a particular purpose and both of them have different advantages and disadvantages so descriptive research is a research method used to observe and describe things without having to specify or conclude um a causal relationship right we're not looking for the cause of the thing that's happening we're just looking at what's happening that's it in experimental research the point of it is to demonstrate a cause and effect relationship between two variables and of often times in experimental research variables are manipulated and controlled you're looking for cause so take my honeybee experiment am I trying to find out why honeybees spend longer time on squashed flowers than they do on lavender flowers I'm interested in that but I'm certainly not going to be able to find that out from just observing them right I'm I'm not going to know why I'm just going to know if they do spend more time on squash flowers than lavender flowers but I'm not going to have any clue why I'm not manipulating anything I'm not um doing an experiment I'm just I'm just watching what's already happening to see what's going on um so my research is descriptive here is a helpful picture to help you differentiate between descriptive research and experimental research and again we are going to go into so much detail about these different types of descrip itive research um and how they are so different from experimental research but this table you can go to save it it'll help you on your tests and stuff um but it'll help you differentiate between the two and so under descriptive research we have case studies naturalistic observation surveys and again the purpose is to observe collect and record data um some advantages of descriptive research is it's good for early uh for developing early ideas um it's more helpful when we're just trying to reflect on actual behavior um than other methods and it's easier to collect data because we're not conducting a full-blown experiment where we have to manipulate the environment and the variables the disadvantages of descriptive research is there's little or no control over the variables right we're just watching things as they're already occurring so we're not actually controlling the variables like I can't go out there and control the bees there's just no way I'm just watching what they do naturally um and then obviously there is no proof of cause and effect so there's no way I'm going to know why bees are spending more time on squash flowers than lavender flowers I won't know based off of my research because I'm not conducting an experiment and I'm not controlling the environment and the variables in order to be able to prove cause I'm just watching what's already occurring so that's those are the disadvantages of descriptive research now experimental research um includes and involves manipulation and control of the variables and the purpose is to identify cause and effect we want to know why things are occurring not just watch them occur and be able to describe how they're occurring and when they're occurring and but we're going to want to know what's causing them to occur and that's that's when you when you have to conduct an experiment the advantages of an experiment is that it allows researchers to precisely control the variables and to identify cause and effect which is kind of the point it's just a really complex way to go about finding information out costs a lot of money too the the disadvantages of experimental research are there's a ton of ethical con concerns um there's practical limits artificiality of lab conditions um you have to be really careful when you're conducting an experiment especially when you're involving human participants right because that's usually um involving manipulating their environment and manipulating variables that involve them and so you have to be really really careful in an experiment the different types of descriptive research are naturalistic observation surveys case studies and correlational studies so natural naturalistic observation is super easy because you're just watching things happen as they're already occurring um so re researchers watch while people or things right I'm watching honeybees behave as they normally do um and we'll talk a little bit more about each of these types but um that's what naturalist ISC observation is so my my um research study that I'm conducting is naturalistic observation um next we have surveys so the researcher asks people a series of questions and it can be done in person or by telephone email or written questionnaire case studies are super interesting because it focuses on like one single person maybe two like if it's a couple or siblings um medical and psychological practitioners who treat people with problems often conduct case studies to help determine whether therapeutic interventions produce changes in their client symptoms so case studies are super interesting and then we have correlational studies which at first I feel like students feel they struggle to understand what the difference between correlational and experimental is but we'll talk all about this correlational studies research how two things are related to each other um and they're super super useful okay so we're going to go into detail about all the different types of descriptive research um and what it is exactly and the pros and cons of using each so the first one is naturalistic observation and it's really systematic people watching but it doesn't necessarily have to be people the only reason why it says people is because this is a psychology class but naturalistic observation can be used to observe anything like honeybees or animals or or the weather or I don't know how water behaves in the ocean like literally anything the advantages of naturalistic observation are that you can study things that are too unethical to do in an experiment let's say I am really interested in learning about people who murder people I'm not by the way but let's just say I was if I were to conduct an experiment I would have to manipulate people to Mur people because you have to have complete control over your experiment and you have to manipulate and create the environment that you're trying to study obviously that's very unethical so an easier more ethical way to do this is to study people who have already murdered people I'm not making them murder anybody they already did it on their own um and now I'm just observing them maybe in prison Okay so that's one one of the benefits to naturalistic Observation another Advantage is that you can study things that people might lie about um for example let's say you want to know on average how many people adults pick their nose like who's going to say I pick my nose no one no one's get meant to picking their nose but if you're watching them behind a what are they called double-sided window or mirror or whatever um you might see them pick their nose right you're not making them pick their nose they're doing it on their own um so it's a huge benefit you can just watch people behave as they normally would you're not manipulating anything if you're doing it right okay there's something called the observer effect and it's actually a a term in physics but it applies to psychology and it refers to the fact that the process of watching something sometimes can alter how that thing behaves so for example if I'm sitting in my backyard watching bees and I'm too close to them they may get freaked out and not come around right I might be noticing that there's like less bees going on all the flowers because of my presence um so you have to be really careful about that if it's if natur if you want to do true naturalistic observation and get the most accurate results whatever you're observing um the environment shouldn't be altered at all and what whoever or whatever you're observing shouldn't even know you're there that's why I included this picture it's a picture of a woman watching the children and they have no clue she's behind that that window watching um if she was in the room with them they would be aware of her presence and they may act different which would not give you the right results right that's not going to give you the results um you're not going to witness what what actually happens naturally some disadvantages um of naturalistic observation is that it's super timec consuming like I said watching bees for a couple hours seems really boring and timec consuming um and it doesn't determine cause and effect so I'm not going to be able to know why bees spend more time on one flower than the next I'm not going to be able to determine that from my exper or from my research um I'll just know if they spend more time on one flower than the other so that's another disadvantage the next type of descriptive research we're going to talk about is surveys surveys are probably the easiest descriptive research type um you just ask people directly about their behaviors obviously surveys is people oriented right I can't go out there and ask the bees like hey how long do you spend on squash flowers and how long do you spend on lavender flowers um so it has to be people and it has to be people who you can communicate with um and who are going to be able to answer the questions as honestly as they can um sometimes surveys go indirectly or finding information about um a person so let's say you're interested in children right A lot of times children won't wouldn't be able to really answer questions so you might give a survey to their parents um the advantages of surveys is it's super quick like especially these days with technology you can like send out a survey online um you I mean it's so easy to do surveys they're cheap right now because we can just do online surveys you don't have to like print paper um so it's it's very easy and very cheap the disadvantages is that sometimes people don't tell the truth well oftentimes people don't tell the truth especially if you're asking them something that maybe conflicts with like social norms or or might be considered taboo um people will just lie because they're not comfortable being honest about certain things for example like like I said in the previous slide what if you wanted to know how many adults picked their nose most people aren't going to admit to picking their nose if they did you know what I mean if you ask someone a survey do you pick your nose they're going to lie and they're going to say no that's disgusting and then they're going to get in their car and they're probably going to pick their nose um so surveys don't always give you the most accurate truthful honest information which kind of defeats the entire purpose of a survey so you have to be really careful about um the questions that you ask and how you ask them and so that's why a lot of times when you're taking inventories and we'll take some in this class we'll take some personality inventories and like love language inventories but a lot of times the questions seem super ambiguous like you don't really know why they're asking you it what are they why are they asking me this like what are they trying to figure out I have no clue what they're trying to figure out by asking me this that's the point that is totally the point um they just get more honest answers sometimes a lot of um questions will be set up in a way to where you are honest and and it's telling but you have no clue that you're doing it for example let's say you wanted again let's say you're trying to figure out if people pick their nose and if you just like blatantly say do you pick your nose most people are going to lie but maybe if you rewarded it a little differently like um let like a scenario let's say you are you just sneezed and you don't have a tissue and you're by yourself what do you do you know maybe somebody would be more likely to say oh I picked my nose or maybe I'd wipe it on my shirt or whatever you get an you get more honest answers when it's more it's less obvious what they're trying to ask you and then of course another disadvantage which is a disadvantage of all descriptive research is that you can't determine cause and effect so you're not going to be able to know why things are occurring you're just finding out how and if they occur next we have case studies case studies is another type of descriptive research um they are usually an intensive study of one or two people like a couple or siblings but usually it's just one person the advantages of a case study is it's the only method you can use if the type of behavior you're looking at is super rare um so for example like let's say you are you're wanting to investigate why people engage in school shootings um there's no way you could get a large sample of School shooters like that's not a huge population of people thank goodness right we don't want it to be um but instead you would have to do a case study of someone who did a school shooting um people have done case studies on Eric Harris and Dylan uh clbl to try and figure out why they uh why they participated and did the Coline high school massacre um they've talked about how these children were bullied at school and how they played a lot of violent video games and although this may be true you have to be really careful not to generalize these findings to other School shooters um or even just worse like just to someone who might be tempted to say that bullying video games cause the school shootings which of course we wouldn't be able to say just from collecting data from these two these two people so um you know you're getting very detailed information about this person or this these two people um you just aren't you really can't generalize these findings like you can with other types of research like my research for example on honeybees I'll probably be able to generalize them to other backyards in my my neighborhood like oh I bet other people have honeybees in their backyard that spend a lot of time spend more time on squash flowers than lavender flowers you know it's safe to say that but with regards to case studies you know because the population is so small you really can't generalize these findings you're really only finding information out about those that those individuals that you're studying the disadvantages right that you can't generalize the results to all people who fall under that population and it doesn't determine cause and effect right you're not conducting an experiment so we can't for sure say that violent video games and being bullied causes someone to turn into a school a school shooter we can't say that there's tons of people who play violent video games and get bullied and don't kill anybody so this does not case studies do not help us determine cause and effect it just gives us a ton of information a ton of detailed information about one or two individuals and the last type of descriptive research is correlational um and correlational research studies how two things are related to each other um so it's going to determine if something has a positive correlation or a negative correlation a positive correlation is when one variable able increases the other increases for example studying and GPA right there's usually a positive correlation with the two variables of studying and GPA usually when one increases studying right when you study more the other increases GPA when you study more your GPA tends to get higher so there's a positive correlation that's the the correlational relationship between studying and GPA positive correlation or you can have a negative correlation when one variable increases the other decreases for example exercise and weight so the variable that increases let's say you work out more you start working out more and then the other variable your weight goes down and decreases so that's a negative correlation so that's really what we're looking at here with correlational um research the advantages of this type of research is that it describes how the variables are related right so we can say at the end of our research that when this increases this also increases or when this increases this decreases so we'll be able to say how they're related if they have a positive correlation or a negative correlation and that's helpful right it's helpful to know if things are related because it helps us predict how what's going to happen if we notice the first variable increases we can say oh well I bet the second I bet the other variable is going to increase because we already determine that they have a positive correlational relationship the disadvantages of correlational studies is that you cannot determine cause and effect so even though we can say studying and GPA have a positive correlation we don't know for sure if studying is what's causing GPA to increase we have no clue we're not we're not conducting conducting an experiment so we can't for sure say that but we can say that when studying increases GPA tends to increase now the reason why we can't determine cause and effect other than the fact that we're not conducting an experiment is there maybe some third variable in there that we're not looking at we're not even aware of that's actually causing the GPA to increase maybe this semester the person decides to study more often and the GPA is increasing but what we don't know is maybe the classes they're taking that semester are easier so we just there's third variables in there that we may not know about so we can't say that there's that studying is actually causing the GP to increase we can only say that they have a positive correlation a really great example of a correlational study is the one on the slide so this correlational study is looking at stress and depression so those are the two variables they're studying and what they've been able to determine through this study is that they is a correlation between stress and depression a positive correlation when one of the variables increases the other increases so when stress increases depression increases and when depression increases stress increases do we know that the stress is causing the depression to increase no does the depression increasing cause the stress to increase we don't know we don't we don't know all we know is that when stress increases depression increases and when depression increases stress increases that they are related there's some relationship between depression and stress where um they're both they're both increasing right they're they're related they're correlated something that this research is not showing is maybe a third factor for example poverty maybe poverty is what is causing both of these to increase right and so because we don't have that third Factor because this is not an experiment we cannot prove a causal relationship we cannot prove what's causing the stress to increase we cannot prove what's causing the depression to increase is is this saying that stress is not causing the depression well no is it saying that depression is not causing the stress well no we're just saying we don't know we can't determine that from a correlational study because we're not in full control over the study because it's not an experiment all we know is that the two things are correlated in some way they're related in some way and we can tell you that it's a positive correlation because when one increases the other increases So within a a correlational study is something called a correlation coefficient and then this is a statistic used to express the strength and nature of the relationship between two variables and it can range from + one to negative 1 the sign of the coefficient Plus or negative tells us the direction and the number tells us the magnitude or strength of the relationship between the two variables um so you can look at the little picture on the right that tells you the type of relationship two variables have um depending on the correlation coefficient that they're given so now we've finally finished going over descriptive research and we are moving into experimental research and experimental research examines how one variable causes another variable to change so we are looking at and measuring and finding out cause the advantages again it can establish cause and effect and it can eliminate outside influences for example with the correlational study that we just talked about with the depression and the stress and there possibly being a third factor or a third variable called or what did we decide it was poverty right poverty could have been influencing both of those two variables in an experiment we have full control over the environment and the variables and so we can eliminate those outside influences the disadvantages of experiments we may not be able to generalize right because um it's such a specific population and Sample and such a unique situation the experiment that we're creating has has created that we can't generalize all the time to just like the general population that um matches who were experimenting on and sometimes experiments are too unethical to do often times experiments are too unethical to do that doesn't necessarily stop people from doing them they just end up like paying participants um a lot in order to get them to participate now that we're done talking about research methods it's really important for you to pick the most appropriate research method depending on what you want to study this will be on the test um not these exact questions but you will need to be able to determine the most appropriate research method depending on what you're studying so go through these examples the first one is were serial killers abused as children what would be the most appropriate research method for this think about the population is there a huge population of serial killers no I mean I hope not not that we know of do we have access to a huge population of serial killers no and so what sounds like the most appropriate way to research this probably a case study most serial killers are studied using case studies right where just one person's just interviewing them and and collecting data on one person what type of students work out on campus how could you figure this out which which research method could you use to figure out what types of students work out on campus it's definitely going to be a descriptive type of research because you're not proving any causal relationship on why certain students work out on campus you're just trying to figure out what types of students work out on campus so if I were doing this I would definitely do naturalistic observation um you could also do a survey but some people might lie and say they work out on campus when they don't because you know we all have that guilt when we don't work out and go to the gym um so I would totally do naturalistic observation what I would do is I would just act like someone working out at the gym so that I'm not like it's not super obvious that I'm observing people because I think people would become very uncomfortable with that and they may like leave or like try to get away from me but I would just act like a fellow exercis and um collect data the next is does meditating help to reduce stress so this is a tricky one is this descriptive or is it an experiment so it's not necessarily saying does meditation reduce stress right if it was saying does meditation reduce stress that sounds like an experiment because it's it's insinuating that meditation causes stress reduction but because it's saying help to reduce it sounds more descriptive it sounds more like a correlational study does meditating have anything to do with stress reduction so we could totally do a correlational study for that we're not trying to figure out if meditation is the cause of stress reduction we're just trying to see if when you meditate so when meditation increases if stress decreases and and then which type of drug a b or c helps to reduce cholesterol levels clearly that's an experiment because there's a drug involved right so we're completely manipulating the variables here and then do women smile more than men how would you conduct that you could do a survey but I feel like people don't really aren't self-aware enough to know how much they smile so I would do naturalistic observation something that's really important to include in an experiment is that you have two groups you always want to have an experimental group which is a group that's actually getting exposed to the manipulation or the treatment in your experiment and then you want a control group um and this is the group that gets no manipulation or treatment and it's used to compare how the independent variable changes the dependent variable um and it helps it just helps you be able to rule out other things because you have a group of people who are not receiving the treatment um and so let's say that you have you pull a sample of people who have high cholesterol right that's your population that your that's your sample of your population your population are people who have high cholesterol you pull 100 people from that sample or from that population to create your sample of people with high cholesterol and then you randomly assigned half of those people to the experimental group who gets the actual cholesterol medication that you're trying to see if it works and then the other half would be the control group and maybe they don't they so they don't get the cholesterol medication they get a placebo pill so they don't even know know that they're in the control group and the people in the experimental group have no clue they're in the experimental group everybody's getting a pill they just don't know if it's the placebo pill or not this is important right because if the experimental group who gets the actual cholesterol medication let's say we notice that their cholesterol levels lower and then the control group cholesterol levels also lowered but we know they didn't get the cholesterol medication so that does not prove that the cholesterol medication worked because for some reason the control group also had similar results even though they were not receiving the treatment so it's really important to have a control group if that experimental group showed that their cholesterol lever levels decreased and the control group cholesterol levels did not decrease they stayed the same it's a very good indicator that the cholesterol medication worked so how do you decide who goes into the experimental group and who goes into the control group it has to be random it just has to be that's the only way to to accurately conduct research so the researcher should randomly assign who goes into each group a participant with characteristics that could influence responses will have an equal chance of being in either group so everybody from your sample of 100 people with high cholesterol each person has has to have the same chance of going into the experimental group as everyone else um and this should make the group similar before the study Begins the groups have to be as equal as they can be you can't cherry-pick who you want in your experimental group right like let's say you're trying to um sell and produce a weight loss drug right and so you pick a hundred people who are overweight and are volunteering to try out this weight loss drug and you cherry-pick people who look like maybe they would lose weight easier to be in your experimental group who are going to be getting the weight loss drug maybe these people were asked to survey and they they have better or healthier eating habits so you cherry-pick them and you put them into your experimental group because it's going to make your drug look like it worked that's not fair and you can't do that and this is why people are held to the scientific method so that you can't cheat because B people will and especially with drug companies and pharmaceutical companies they want their products to work and they want to have data that shows that it works so that they can sell it and make money so this that's why it's so important that people follow the scientific method and these very strict principles um that align with the scientific method something that also helps a avoid biases and avoid cherry-picking of who's in what group is to do a double blind procedure um and this is this means that neither the participant or the researcher knows who's in each group obviously there's a list somewhere that someone has that that states which group each participant is in but the researcher who's interacting with people with the participants don't know and that's really important because if you know something about someone you might treat them differently which may alter the results right so the per the researcher who really wants this weight loss drug to work might if he knew who was in the experimental group and who was in the control group he might treat the people in the experimental group differently or might say things differently to them that encourages weight loss or encourages them to be more active um so it's really important to do a double blame procedure where no one knows that way there's no um there's no difference in treatment um to to the participants so that all the participants are treated to and talk to the same so you don't have to do this practice but if you want you can do it um a researcher wants to know if taking the pill blimpo helps reduce unwanted belly fat he randomly assigns 50 people to take blimpo and 50 people to take a sugar pill which is a placebo pill he doesn't know who's in which group and the participants don't know whether they're in the experimental group or the control group why would having this double blame procedure help to control bias in this study well I just explain that to you but think about that and this applies to a ton of different types of experiments not just like weight loss not just drugs but just a pretty much all experiments once experimenters have collected their data they have to make sense of their results and they Ed statistics to do this no this is not a statistics class if it was I would definitely not be your teacher because I despise math however statistics are so important to the field of psychology because it helps us make sense of our data when we collect data and it's applicable and it tells us amazing things about people and why we are the way we are the definition of statistics is a mathematical method used to report data and to understand and and make sense of data descriptive statistics describe and summarize the data and it's typically using measures of central tendency which should be a refresher for you if you've taken like high school level math so we're looking at mean median and mode um we also look at standard deviation and inferential statistics and we'll talk a little bit about this on the next slide measures of central tendency are the mean median and mode um the mean is the average value out of a set of scores so you add up all the scores and divide them by the total number of scores so let's say I collected I collected the time spent on squash Flowers by 20 bees right so I watched 20 different bees land on a squash flower and I measured how long they were there and remember I want to find the average time bees spend on a squash flow and compare that to the average time honeybees spend on a lavender flow so what I would do is I would take up all the T times like there's 20 different set there's 20 sets of times from different bees who landed on squash flowers like maybe it's ranging from 1 second to 2 minutes so I'm going to take all of those 20 times and I'm going to add them up and then I'm going to divide them by 20 because that's how many um bees I collected times for and that's going to give me the average time the honeybees spent on the squash flower next we're going to talk about the measure of central tendency called the median and that's the value in the middle or the midpoint so what you would do to figure out what the median of some of a set of scores or times would be was you would put them in numerical order from smallest to largest you would count the total number of scores and if there was an odd number of scores or times it's the number in the middle so for example if I had only collected times from five BS right that's an odd number the median would be the third time after I put them in order um let's say that the smallest amount of time was 1 second the second was 30 seconds the third was 45 seconds the fourth was a minute and then the fifth was a minute and a half the median of those scores would be the 45 second I think that's what I said I don't even remember what I'm saying if this isn't making sense there's an example on the right that you can use now if there's an an even set of scores or times like in my original experiment I was collecting times from 20 B's that's an even number so there there is no number in the middle what I would do is I would take the two Central scores and find the average and that would be my median and then the last central tendency we're looking at is mode and that's the value that occurs most often so I always the way to remember this is mode kind of rhymes with most so it's the number that happens the most um and so like if you look on the right there's an example of set of scores and you're going to notice that the number six happens the most in the first example and so the mode would be six and then the second set there there's no number that happens more than the other so there is no mode I mentioned that standard deviation is something in statistics that's really important to look at in Psychology and I'll tell you why so what is standard deviation standard deviation shows how much the participant scores vary from one another in each group um and this is important um so it kind of helps you to decide and determine um how much scores are deviating from each other because that's an important thing to take into consideration and the example that I'm going to use is one that looked at people who were watching um two different shows so we have a group of people who watched Pokemon and then we had a group of people who watched Extreme Makeover home Edition and after they watched the show they were given the opportunity to make a donation to a charity okay and so the numbers that you're looking at under each show are the amount of dollars donated so if you were just to look at these scores because what we're obviously the researchers is trying to figure out which show made people donate more right was it Pokemon did Pokemon in like inspire people to donate more money than Extreme Makeover home Edition or was Extreme Makeover home Edition more inspiring to get people to make charitable donations so that's really what this researcher is looking at but when you look at the scores and you calculate the mean it's the same mean they're both five the average amount of money that people donated was five and so it it looks like it doesn't matter what show they watched they were the average per the average uh amount donated was $5 but if you actually look at the set of scores you'll notice they don't look the same right you'll notice that on like for Pokemon the scores don't deviate from each other very much you have four and five and six and five and four and six they're around the same amount they're not deviating very much from each other but then when you look at Extreme Makeover home Edition and and how much people donated after watching that you see something different these scores deviate a lot from each other a lot like we have a score that's 12 we have a couple ones we have an 11 and that's an important factor to look at right so yeah the average donation was the same but clearly Extreme makover Home Edition had a different impact on people at least some of them than Pokemon did Extreme Makeover home editions seem to inspire people a lot to donate way more than Pokemon did now it didn't Inspire everyone to make that same kind of donation but it it inspired a good amount of people to make to feel super inspired to make a bigger donation so standard deviation is important because it tells us more about our numbers the S the measures of central tendency don't always tell us a whole lot it gives us like the mean but again look at the look look at the mean for both of these shows it's the same when the shows clearly didn't affect people the same way so the mean didn't really show us that we need to look into it a little deeper a little further and and look at the standard deviation to see how much the scores deviated from each other I think something that a lot of people are curious about when we're talking about experiments and research that's conducted um there's a lot of concerns about what ethical research guidelines olist follow um and anytime someone wants to conduct research in the field of psychology they have to submit it to the IRB for example when I was getting my master's degree and I had to do a thesis which required me to conduct research I had to submit my research um my research plan to the IRB because they're going to make sure that I'm following ethical guidelines before I begin my research and if they determine that I'm not I'm not um following these ethical guidelines then they don't allow me to do my research so I have to get I have to get their approval before I anybody has to get their approval before they conduct research um and some of these guidelines you have to follow is you always have to obtain informed consent so anytime you're participating in any type of research you have to sign you have to give them consent to um which is basically your permission to participate in the study and you have to at least be informed about any risks or benefits of being a participant they may not be able to tell you in informed consent what they're measuring or what they're researching but they at least have to tell you the risks that you're taking by participating in the study and this has to they have to obtain the researcher has to obtain informed consent prior to you participating in the study another guideline is protect participants from harm and discomfort people are not supposed to be put in situations where they're going to be hared or hurt or uncomfortable um I mean they they shouldn't be traumatized it shouldn't be something that's going to have lasting negative effects um and then also the researcher has to protect the particip participants by um using confidentiality so their personal information cannot be shared maybe um instead of using their names they give assign them numbers right so that's a way to protect the participant and then another guideline is they have to provide complete debriefing after the research takes place so once it's over with they need to reveal to the participants any information that was withheld during the study like okay what you just experienced was this because we're researching this right a lot lot of times you can't say that ahead of time because it'll alter the results if people know what's being researched they're going to respond differently and they may behave differently um and so it's better that they're kind of left in the dark so that they can answer and behave more honestly um it's going to give more accurate results and so um the rule of thumb for that is that you at least have to tell them once it's over what really was going on something that I'm always concerned about because I'm an animal l is are there any like ethical guidelines that researchers have to follow regarding animals and so there is for the most part um the institutional Animal Care and use committee approves or denies permission for studies involving animals um and what they're looking at is they want to make sure there's evidence that the research is going to not involve harming or um yeah harming any animals so the animals should be protected from Pain there should always be alternatives to using animals looked at first like if there's a way to conduct This research and get and find out the same thing by not using animals then that should be used before using animals and what's the clinical and scientific Merit of this study is it worth it is it worth hurting or using animals to find this information out and that's a very great area right because animals are used in a ton of different scientific research that I personally don't find um worth it for example Cosmetics like I just don't think animals should be included in testing of Cosmetics Cosmetics are not um saving lives they're not like a requirement to stay alive like we don't need Cosmetics it's a privilege to have Cosmetics it's a choice to have cosmetics and this is coming from someone who wears makeup and used to work at a makeup counter um but I just don't think it's worth testing animals over I really don't so but I mean it's still allowed people still research companies still use animals in testing of cosmetics and beauty products so that's a gray area for me um I personally don't feel like that's right but it's still being used in our country um you can totally avoid this or um not go along with it just by maybe finding and buying products that are not tested on animals so just in case you were wondering about that part of the research