that's why it's the one that everyone's been asking for it's a full breakdown of research methods you can see the OCR spec that we're going to cover over the course of this video it's a little bit too complicated though so I've simplified it down to just this list of stuff that we need to know and over the course of this video we're going to cover every single one until you're an absolute expert in research methods let's not waste any more time and get straight into it firstly it's important to remember that research methods can be tricky and you might not get everything the first time round for example we might refer to types of validity later on when we talk about experiments this means if you didn't quite get validity you might have to go back and rewatch it likewise I'll be trying to keep everything really brief and simple and that means you might have to watch something a couple of times for it to go in anyway let's get on with variables now these are the only four variables that you need to know the independent variable is the thing that researchers are changing or manipulating in an experiment now when teaching research methods I always like to use a little study that helps me to remember how to apply stuff and this is the one on screen a researcher wants to find out if temperature affects math scores and this is how I do it so I take a group of students I'd split them in half half would do a math test in a hot room half will do it in a cold room and then we'd work out whether the hot or cold room was better now remembering that an independent variable is the thing that's being changed in an experiment can you apply that knowledge to this study what is it in this study that is being changed pause the video and have a think the answer is temperature so the thing that's being changed by the researcher is the temperature we're putting people in either hot rooms or cold rooms now the dependent variable is the thing that we are measuring so as a researcher we're measuring something and that is called a dependent variable now let's bring it back to that study again what are we measuring have a pause and it's the math scores knowing what an IV and DV is is super important for later on when we talk about hypotheses now anyway let's move on to co- variables and these are basically the same as an IV and DV except that these are used for correlations and we'll talk about correlations a little bit later on but if you ever get asked to do a correlational study instead of saying we have an IV and a DV we'd say we have co-variables instead finally we have an extraneous variable and these are variables that might affect the results of the study which the researcher hasn't chosen to control so for example in the hot or cold room condition it might be that the hot room there was a really quiet room in the cold room maybe there was loads of noise going on outside that would also affect math scores and it would be an extraneous variable a hypothesis is a prediction and every piece of research that's ever been done in Psychology has had a hypothesis now in order for us to be nice and scientific we have to make sure that we're writing a hypothesis in a correct way and that's what you're going to have to do for your exam to be able to write a hypothesis either as an alternate hypothesis or a no hypothesis these are the two terms you need to know an alternate hypothesis predicts that there will be a difference and you can see a template on the screen and if you can remember this template whenever you get asked a question about an alternate hypothesis you'll be well on your way to some marks now let's go back to that study and try and apply it to the research we've had before pause the video and try and fill in the gaps for me and the answers are coming up now that's how you write an alternate hypothesis there will be a significant difference however on the other side we've got a null hypothesis this is the same but it's simply says that there will be no significant difference so a nor hypothesis is a prediction that says that there will be no difference I know I'm going really quick but there is so much to cover but if you are finding it helpful please make sure to share it with others in your class types of data next and there are four that you need to know we'll get started with qualitative and quantitative data qualitative data is simply anything that's sort of a longer written response so like an interview transcript or newspaper article that would be qualitative anything quantitative would be numerical and to give you an example of this it would be like math scores in a test just like the one that we did in our study earlier on then we need to know primary and secondary data primary data is any data you've collected yourself secondary data is any data that you've gathered from other sources for example maybe from an article online or a book you also need to make sure that you know the strengths and weaknesses of these types of data with quantitative data because it's numerical it's so easy to compare for example with numerical data like in exams you can compare when you got 90% correct with when you got 60% correct and you know that the 90% was better than the 60 that's numerical data and it's so easy to compare whereas if someone said to you you did really well compared to very well well that would be qualitative data and well it's really hard to compare so that's a strength of quantitative data a strength of qualitative data is that you get tons and tons of detail compared to quantitative data so you're really able to try and understand why a certain Behavior happened with primary and secondary data primary data is great because you can trust the data you've collected it yourself you know that it's to a good level and it's the exact stuff that that you want to find out so primary data is great for that reason a strength of secondary data is that you can save a load of time a load of money and you might be able to access data that you wouldn't otherwise be able to access such as information in other countries or areas around the world next we come on to sampling methods and this is how psychologists gather their participants for the research the target population is the type of person that we're going to try and aim for so for example a research wants to find out if temperature affects math scores in year 11 students in the UK the target population is every UK year 11 student the sample is the specific group that we gather so it might be the 30 students in a class that I have access to that I will use in my study that's called a sample now in order to gather a sample there are three ways that you need to know random opportunity and self- selected a random sample means that every single participant in that Target population that is every single year 11 student in the UK has an equal chance of being chosen an opportunity sample is just whoever's there at the time so for example I might choose uh the students in my class that I have access to or a self- selected sample and this one's a little bit tricky is when they volunteer for it and it's important that you don't say something like I would just walk into a room and get them to put their hands up they have to step forward and the easiest way to bag some marks in an exam when you get asked about a self- selected sample is to say something along the lines of I would put up posters around that place and wait for them to respond with my phone number and my email on it so in this case I would put up posters around a a school and I would ask for year 11 UK students to respond the problem with this of course is that only a certain type of person might step forward for a piece of research this means that the sample might not be representative a fancy word for saying that it doesn't represent everyone and that's a problem in research now with experimental designs now that we have our participants we need to be able to use them and allocate them we have two ways of doing that that is an independent measures design or a repeated measures design an independent measure design would mean that all participants take part in just one condition so for example there would only be an either the hot room or the cold room with a repeated measures design all participants would take part in both conditions so they do the hot room in the morning and then maybe in the evening they do a math test in the cold room this is a repeated measure design and as you can imagine there are some strengths and weaknesses of each you see a problem with an independent measures design is we have something called individual differences what if the group that I put in the cold room just happened to be better at maths than the one in the hot room this would affect the results big time however they don't suffer from something called order effects which is a problem with repeated measures with repeated measures although we don't have individual differences because think about it every participant's in the hot room and then they do the cold room and you're just comparing the results to themselves at the same time they will have something called order effects this means that they might do better in the second test because if you think about it they might be warmed up to the idea of doing a math test or maybe they haven't woken up properly in the morning test then they do better in the second one because they've woken up a little bit more either way this is an order effect and this is one potential problem of a repeated measure design you do need to know definitions and strengths and weaknesses next reliability and these are really tricky so please make sure you're paying attention to get all of these key terms whenever you see the word reliability I want you to think about consistency and replicability how much can I repeat a study and get the same result that's reliability and there are three types that you need to know so the first type is internal reliability and this is basically how much the study would be reliable within itself and I know that sounds a bit confusing but bear with me let's say we're doing the hot cold room math test if I was to get students in the hot room to do the math test three times and the first time they got an A star the second time they got an F the third time a c well that wouldn't be very consistent and that's low internal reliability whereas High internal reliability would be if I got them to the test three times and all three times they got a stars or they got the same score next we have external reliability and this refers to whether it's consistent at a different time or a different place for example if we're doing a math test here and then a math test in a different school when we get relatively similar results we could assume that it would be high in external reli relability because they're matched up in a different situation so internal reliability is all about within the study external reliability is across different studies or different places or different times finally we have inter rator reliability and this is basically how much different raters or different people may be observing or scoring the same research agree on a result for example if I had three people marking these math tests and all of them agree on the same scores that has high inter rate reliability if the three of them came up with completely different results that's low inter rate reliability we tend to use this when it comes to observations so if three people are observing a certain behavior and they all agree on the same sort of scale that would be high into rate of reliability this is a good one to remember for Williams atal study in sleep and dreaming we then come on to validity and think of validity as how accurate it is so if reliability is replicability and consistency well validity is all about accuracy is it correct and there are three types that you need to know the first is ecological validity this basically means is the research done in a realistic environment is the setting real if so then it has high ecological validity if it's done in an artificial lab study then that would be low in ecological validity we then have population validity and this is basically saying is the sample representative of the entire Target population if I wanted to do the hot and cold room math test and I ended up picking um 10 of the absolute best top set math students well that wouldn't be very representative of every student in the UK likewise the students at my school might be very to students at your school and at other schools around the world this would be low population validity High population validity would be a huge sample with loads of backgrounds and loads of different places around the world taken into account finally we have construct validity I think this is the trickiest one this refers to when a study tries to measure something really complicated but it does it in a way that's maybe a little bit too simple for example in Blackwell we're trying to measure mindsets but we're doing it with just a questionnaire in heaven study we're doing delinquency with a questionnaire as well so we're measuring stuff that's really complicated and we're doing it in a really simple way this means it has low construct validity so this is when psychologists are not measuring something in its entirety next we come on to types of biases and this is basically just a big key terms list as we go through it'd be really helpful if you're trying to think about some of the studies you've encountered in your time doing psychology and try to apply them for example the first one gender bias this is when there's an uneven or unrepresentative amount of the genders in a piece of research for example in Williams atal study into the Bizness of Dreams there were 10 females and two males this would be unrepresentative and demonstrates gender bias culture bias is similar but rather than it being about genders it's about the culture and countries that are involved for example a study like Blackwell made use of just the American education system to decide that growth mindsets improved your academic achievement in reality that's just in America it's culture bias because we can't say that all education systems in the world would react in the same way next age bias is when a researcher favor certain age groups over other age groups for example Cooper macki study into video games and agression only use children from the fourth and fifth grades in school in America well this is age bias because we can't accurately say that these results could apply to all ages all children only the ones in this age range next we have experimental bias which is when an experimentor might manipulate some of the research in order to support their own theory for example it could be argued that P is experimental bias and this is because he's used his own study to support his own Theory and even use his own kids in it Observer Bias happens when a researcher's beliefs might affect what they observe instead of recording what's really happening their bias might influence what they're noticing or how they're interpreting it and therefore how they report it this leads to Observer Bias finally we have bias in questioning and this is when you phrase a question to support one Viewpoint more than another for example you might try and manipulate the way that somebody answers a question instead of saying where were you on the night of the 20th you might say why were you at the cinema on the night of the 20th this suggests subtly that that person was at the cinema and that demonstrates an example of a question with bias next we come on to ethical guidelines and these are so important when carrying out psychological research this is because it's really important that psychology has a really good reputation so we can hire more participants and do more research and learn more things about human behavior so when doing research we must follow six really important rules and these are on the screen now the rules are fully informed consent this means that participants must be told about the full truth about an experiment and what they're signing up for and they must agree next we have deception now this is something that is a little bit questionable when it comes to psychology deception means lying to a participant and we should keep that to an absolute minimum sometimes it's kind of unavoidable though next confidentiality participants should remain as Anonymous as possible Sometimes using code names is a great way to keep people Anonymous next debrief studies should end with a debrief this is just a small chat to make sure that participants are psychologically mentally physically okay and that they're leaving the experiment in a good state next we have right to withdraw and this just means that participants should should have the right to leave the experiment at any time and finally we have protection from harm this simply means that participants should leave a piece of research in the same physical and mental state that they entered in now you may be thinking God that's a lot to remember how am I going to remember all this well there's an easy little way of remembering it can do can't do with participants now I didn't make this up and I'm not sure where I got it from if I'm honest so if you do find out please let me know cuz it's absolutely amazing and I like to credit whoever's come up with it but can do can't do with participants is a great way of remembering the six ethical guidelines as you can see the first that of each word stands for one of the six terms if you can remember this when you get a question about ethics in the real exam you could write can do can't do with participants on the side of the page and it should bring back to the top of your memory all of these key terms next we have types of experiments also known as experimental methods now the difficulty here is getting them mixed up with experimental designs so in an exam if you get a question about experimental designs please do not get them mixed up with experimental methods and viice Versa otherwise it can cost you a load of marks there was one pass paper where this would have cost you about five marks worth of issues so these are the three that you need to know we're going to get started with lab studies and these are very controlled experiments where the researcher directly manipulates the independent variable always in a really controlled environment for example maybe you control the noise or it's just like in our study you control the temperature that is a lab study in a field study this is actually done in a natural environment rather than a controlled environment however the researcher still manipulates the independent variable for example maybe it's done at school in a student's normal lesson however the independent variable is still manipulated there's a great example of a field experiment that I'll link below which is a Volkswagen advert where they try to change some stairs into musical piano stairs to try and see if people were more likely to take the steps rather than the escalators I'll link the video below so you can watch the video yourself finally there's a natural experiment this is an experiment where the independent variable is not directly manipulated by the researcher because it's predetermined so for example maybe it's comparing people with depression versus people without depression this means that it's a naturally occurring variable something that the researcher can't manipulate they can't give someone depression and then take their depression away and then check the differences they wouldn't be able to do that it could also be something that's naturally occurring that the researcher can't control interestingly natural experiments can therefore take place in Labs or controlled settings but most of the time they are done in natural environments an example of a natural experiment that you need to know is Williams atal study in sleep and dreaming as always you also need to know the strengths and weaknesses for example the strength of a lab study is that they are super duper controlled this means that extraneous variables are eliminated or at least reduced and this means that they are much more reliable than field or natural studies also they're really standardized this sounds really fancy but this just means that we try and make sure that pretty much all the participants get a very similar experience the word standardize is a fancy word that if you're using an exam it's sure to impress any examiner a weakness however of a lab study is that well they tend to to suffer from something called Low ecological validity we've already spoken about that in the validity area but it simply means that they have artificial settings and this means that they don't reflect a real life environment and so people are likely to act a little bit differently or in an artificial way a strength of a field experiment is that because they're in a real environment they have high ecological validity this means that participants are more likely to Act Naturally however it means that there are loads of extra variables that might have affected the results finally with a natural experiment they can be a little bit more ecal this is because nothing's being manipulated for example you can compare people that already sleep badly with people that do not rather than trying to deprive someone off sleep on purpose a weakness however is that it might be difficult for researchers to match up the participants that they are comparing if the independent variable is already in place for example if we are doing this one about sleeping badly it could be that people who sleep badly just happen to be older than people who do not and then this becomes an extraneous variable and there's all sorts of issues so natural experiments are not perfect just before we move on to interviews let's try and apply all that we've learned to that Source a researcher wanted to find out if math tests are affected by temperature I want you to think to yourself what kind of study would it be if we're trying to bring participants in from a different place into a controlled environment and do this study I'll give you a second to think and it is a lab study because we're inviting them from outside of their natural environment what if I went into a school in the middle of a class and I did it right there that would be a field experiment because it's in a natural environment and I'm manipulating the independent variable before we move on it's worth you pausing it and just thinking about some of those strengths and weaknesses why would I do a lab study or why would I do a field study here next we come on to interviews which are super simple an interview is a self-report method this means that the participant tells us themselves about their thoughts and feelings now there are two types that you're going to need to know structured and unstructured a structured interview just means that the researcher has actually created these questions beforehand they've preset the questions and they know what they're going to ask every single time so every participant gets the same set of questions an example of a structured interv would be a job interview where every participant needs to have a fair champ so that means that all questions would really have to be the same an unstructured interview is one where the participants get asked questions based on their responses so for example like a chat show host so they might be asked to expand on things or they might be asked how did that make you feel or why did you do that if the researcher is expanding on answers then that is an unstructured interview now there are strength and weaknesses of each that you need to know and a great thing about an interview is that we can access thoughts and feelings which we can't really do if we're observing someone or experimenting on them by asking them directly how do they feel we can access their thoughts however there is one big issue and that's social desirability this term simply means that people are likely to lie to make themselves look better and I mean that's human nature people will try and make themselves look better if you were to ask somebody how many crimes have you committed it's unlikely they'd ever tell the truth so doing observations or experiments on them instead might come out with a better outcome now questionnaires are another type of self-report method meaning that we are accessing people's thoughts and feelings by asking them themselves a questionnaire is usually written down and it's a set of questions they can be open or closed and they can also make use of something called rating scales and in fact they're the three key terms we need to know so a closed question is a question with a fixed set of responses they could be a multiple choice answer such as a yes or no or they could be predetermined categories that you have to respond to and it's really important that if you ever get asked to write a closed question you must make sure you give these fixed choices don't just say have you eaten today and expect that to be a closed question because it's a yes or no the reason for this is because technically somebody could respond to that with yeah I had some serial this morning and then after that I and you get the point so you have to make sure that you're giving the participants a fixed set of responses I've seen loads of students miss out on marks in the exam because they forgot to do this another way to make it a close question is by using something called a rating scale and Blackwell used a rating scale in their study of growth and fixed mindsets at schools using a rating scale is an example of a closed question in contrast an open question is one where the respondent can create their own answers something that's more long and in depth for example to what extent do you think that a healthy diet is important in your life asking this question really forces the respondent to make a bit of an elaborated response something a bit longer that gives a bit more detail now why would you bother doing closed or open questions well closed questions give you quantitative data which is great it means that it's numerical and we can compare data really easily like the example I gave you earlier open questions give you qualitative data now if you need a reminder of the strength and weaknesses of those go back to types of data the third section in this video now we have one more major weakness of a questionnaire and just like interviews they suffer with social desire ability it simply means that people are likely to lie because they want to make themselves look better however we can usually gather quite a large sample really quickly because we can just dish them out to loads of people all at the same time so that's one strength they can also be anonymous whereas interviews can't really because interviews tend to be face to face let's move on to observations this involves the researcher watching the behavior of participants there are six key terms you need to know these come in little pairs so we have naturalistic or controlled observations and naturalistic observations are done in a real life setting whereas controlled is when particip are invited into an artificial setting that has been set up for example a controlled observation might be if I want to observe how a child interacts with a mother I might ask that mother to bring that child into a controlled lab in which I can observe their behavior and look out for whatever it is that I'm looking for a naturalistic version of this would be that I go to a nursery and I observe the behaviors and attitudes of mothers and babies secondly they can be what's called cover or over a covert observation means that the participants do not know they are being observed overt means they do know they are being observed an easy way of in cover usually is through cameras finally it can be a participant or a nonparticipant observation now weirdly a really common mistake I see here is that students think that non-participant means there are no participants uh please don't get this mixed up a participant observation means that the researcher gets involved in the observation a nonparticipant means the researcher does not get involved in the observation now I don't know if any of you have heard of vsource it's a really big YouTube channel they've got a great video on something called Conformity which is where people copy other people and in this video Michael from resource the main person he actually wants to observe people conforming to really bad jokes it's a really great video to show an example of an observation that's actually quite funny I'll leave it linked below that's an example of a participant covert controlled observation because it's in an artificial environment where people have been invited in it's covert because the participants didn't know they were being observed and it's participant because Michael himself actually joined in with the group they're observing almost pretending like he was part of the group it's important to remember that different types of observation can be combined so for example you can have a naturalist overt non-participant research or you can have an observation that is a controlled overt and participant observation so have a think to yourself you know when someone comes in to observe your teacher what kind of observation is that I'm going to tell you the answer in three 2 1 and that would be a naturalistic over non-participant that's because it's in a natural environment they know they're being observed and the researcher isn't getting involved with the actual action themselves they're just standing back and watching now each of these six types has a strength or weakness so I'm going to absolutely Bree through them very quickly with naturalistic you've got problems with extraneous variables and that makes it really hard to establish cause and effect controlled observations make things artificial so that means people won't Act Naturally which causes ecological validity problems over observations means social desirability is going to hit in there's going to be this observer effect where participants will behave differently because they know they're being observed coover observations can be seen as unethical participants don't know they're being observed and that can cause all sorts of ethical problems again go back to the ethics section if you need help on this with a non-participant observation the psychologist might miss importance details because they're not actually involved in what's being studied whereas with participant observations it can be really hard for a psychologist to actually measure everything or to be objective because they're too busy getting involved in what's going on now I know there's a lot to remember but I hope you agree it's not too complicated so all it takes is repetition and lots of practice case studies are super simple they involve all sorts of methods like interviews and observations what we want is loads and loads of detail about one person or a very small group of people case studies are often carried out on individuals that are in unusual circumstances for example Clive wearing is a case study you need to know from memory or Freud's The Wolfman both are detailed case studies into one person that usually involved loads of multiple methods over a long period of time to try and figure out the most they can to try and hopefully explain that behavior a strength is that we get loads of detailed qualitative data which means we get loads and loads of detail however a weakness of a case study is that while people are often really unique that are being studied and therefore we might not be able to generalize to other people in the case of Clive it's such a unique case we probably will never see a case like that ever again so how are we supposed to generalize those results to the general population finally we move on to correlations which I would argue is one of the trickiest bits of GCSE research methods it's so easy to get this mixed up correlations measure two co-variables to see if there's a relationship between them so unlike experiments where we've got an independent variable and a dependent variable we've got co-variables one isn't measured and one isn't manipulated we just have two variables and we want to see if one goes up does the other go up if one goes up does the other go down we're measur measuring a relationship for example we might do a study to look at the relationship between happiness levels and the amount of EX size someone does now when a correlation is carried out there are three possible outcomes a positive correlation negative or zero correlation and a positive correlation means that as one thing increases the other one increases so for example as the amount of exercise increases that someone does then maybe their happiness goes up that would be a positive correlation we can also do a negative correlation and this is when as one increases the other one decreases so for example if we found that the more exercise someone does the less depressed they are that would be a negative correlation finally a zero correlation simply means that there's no relationship so if we were to plot things on a graph it would be all over the place now in an exam you can also get asked to write a correlational hypothesis this is exactly the same except instead of saying there will be a significant difference we would say there will be a significant relationship by saying this it's a correlational hypothesis now you need to remember that a correlation is great because they allow psychologists to carry out uh an investigation of behaviors that couldn't be experimented on for example it wouldn't really be practical or ethical to force someone into doing more exercise or less exercise or to take some sort of pills to make them happier or less happy by doing a correlation instead we are able to investigate this without actually manipulating those variables however the biggest problem is that we can't establish cause and effect please remember this it's so important this means with a correlation we can't tell which variable causes what for example is it that people exercise more because they're happier or do people get happier because they exercise more well we can't work that out doing a correlation we just know that as one variable goes up the go up now this is the most important part of this video and if you've made it this far congrats because I'm about to give you the best bit of advice firstly there's loads of stuff in this video so it might feel overwhelming but the best way for it to go in is to practice so go back through this video and every time that I use that source with the temperature and math scores I want you to change them to one of those two at the bottom instead that's great practice so going back looking through the video and every time that I mentioned that Source pause it change it to one of those two other sources and see if you can apply correctly feel free to drop a comment if you're unsure on the answers as I'm happy to run through them with you the second way is to do p paper questions I know that sounds really obvious but doing P papers and giving them to your teacher to Mark is by far the best way of practicing finally as always I end with a blocket and in this case I've done a massive blocket with dozens of questions that I'll keep adding to so scan the QR code and try this fun game out in order to test your knowledge and with that we end this monster of a video that is all of research methods done in under 28 minutes absolutely crazy and I hope it's helped it has taken me hundreds of hours to put this together so I'm really hoping this helps some of you out thanks for watching and I'll see you guys in the next one