Transcription
Hello, students, and welcome back to our next lesson in our AS Psychology course. Today, we'll be looking at using different research methods.
So, conducting research in the field of psychology: there are lots and lots of different experiments going on all the time, and we're going to have to be using different sorts of methods and techniques in order to get the, um, results that we want from these different experiments. So, there's going to be different ways to collect the data, and each of these different sorts of methods and techniques are called research methods, and they're going to have their advantages and disadvantages.
First, we'll start with a laboratory, or a lab, experiment. This is an investigation in a controlled environment where there is a direct manipulation of the variables. So, it's going to measure the dependent variable versus the independent variable. Its advantages are that you can easily control all the variables because you are in a controlled environment where only you manipulate everything going on. As a result, there's going to be sort of less errors and confounding variables, which you can then control. The controlled conditions also make it more replicable and more valid. So, you can conduct the experiment many times in order to try and get an average, and you should always get the same result because you are in a controlled environment. Also, it is much easier to decipher what the cause and the effect was because there are no other sort of extraneous or confounding variables that could mess with what the, um, cause and effect really is.
However, it does have its disadvantages. The first one being that it lacks ecological validity. And what this means is that, um, ecological validity relates to really how, um, like a real-life situation it is. And because it isn't really a true reflection of what, uh, real life would be, we would call that, um, that it lacks mundane realism. And as a result, what we can, you know, what we conduct in a lab may not really be the same things that would happen in the real world. So, as a result, the sort of results that we get from the experiment may not be as valid as it would be if we were to conduct them in the real world where we do have a bunch of confounding variables that would impact the effect of whatever we're investigating. Another disadvantage is that if people are tested in a lab, there may be subject to things we call demand characteristics and evaluation apprehension. Now, what this means is that when people are being tested, they may behave differently or do things differently and not act the same because they know they're being tested. So, as a result, because people aren't acting as their selves, this may not give a true indication of results to what really, you know, was going to be investigating, what the results should be.
Next up, we have a field experiment. This sort of contrasts from a lab experiment because, instead of being in a controlled environment, it is in a natural environment where an independent variable is manipulated to see its effect on the dependent variable. And of course, we can pretty much, um, take a look at the lab experiment's advantages and disadvantages and contrast them. Its advantages are that it has high ecological validity, less evaluation apprehension, and demand characteristics. However, its disadvantages are that it is hard to control variables, people may not give informed consent, and it is harder to replicate. So, pretty much the opposite to what a lab experiment really is.
Next up, we have a natural quasi-experiment. And this is the study of effects of a natural event where the independent variable is, uh, sort of naturally manipulated. So, an example of a natural quasi-event may be the study of the impact on stress or anxiety of people following a volcanic eruption. So, its main advantage is that it's going to have extremely high ecological validity because it is sort of natural observation in a natural event with naturally, in sort of, manipulated variables. So, if you're going to take one word away from all this explanation, it is the word natural. It is very, very, has very high ecological validity. Also, as the research has very little involvement, there are going to be much fewer demand characteristics and evaluation apprehension. So, it kind of seems as if the advantages are exactly the same as a field study. So, as a result, we can, you know, in our minds, using sort of what we've learned so far from the field experiment, also decipher that the disadvantages are that it's hard to replicate and harder to control the variables, and there may be very many confounding or extraneous variables which impact the sort of experiment. Confounding variables and extraneous variables being variables which we have not accounted for in the first place and you can't really control. So, for example, like the time of day. Also, the disadvantage is that there are few opportunities to study these natural events. So, we'll know that a volcanic eruption doesn't come along all that often. So, as a result, it's going to be hard to check sort of or reconduct these experiments again and again.
Another research method is correlational analysis. And this is a technique to see if a dependent variable and an independent variable are associated, and then it also measures the strength. So, if you imagine a sort of scatter graph, that is what correlational analysis is really for. So, drawing up the scatter graph is really correlational analysis. So, the relationship can be causal, which means that there is a definite link between one variable and another; it can be left to chance, which means by some sort of lucky outcome there is an actual relationship when perhaps there may not have been; or the relationship can be due to a confounding variable, which means that an, sort of, anomalous or extraneous variable out there has made there be a correlation when really there shouldn't be in the first place. So, it, of course, is going to have its advantages and disadvantages. The main advantage being that it allows to measure the strength of the correlation, and trends can lead to research in a new sort of different project. So, if we find out that there is a correlation between, uh, number of ice cream sales and number of, sort of, tuberculosis cases, there may be a relationship between the two. So, there will be more research projects in order to try and see if, sort of, tuberculosis is linked towards the distribution of ice cream, which, you know, sounds, uh, crazy, which we'll probably know, but it's just an example. Okay.
Next up, we have its disadvantages, and one of them is that it doesn't show the effect of confounding variables. So, as a result, if there is a confounding variable, it's not going to show how the confounding variable actually affected the correlation. Similarly, we can't tell which variable will cause which, so whether it's the dependent variable which influenced the independent or vice versa. And also, it doesn't state whether the relationship, um, whether the chance relationship that is, is caused by the correlation, whether it was just potluck, or whether, you know, the chance relationship may actually be causal, but we actually, you know, haven't conducted enough research to find out if it actually is.
Next up, we have naturalistic observation. And what this is is that it's observing behavior in a natural setting where the researcher does not influence, um, the behavior at all. So, this can be a disclosed study of naturalistic observation, and this is where the researcher reveals himself or herself to the group; or an undisclosed, um, sort of naturalistic observation where the researcher will remain hidden and observe everyone just from a remote area where they won't be able to tell that he or she is watching. The advantages are that it has high ecological validity because people will be in a natural setting, and also, if undisclosed, there will be no demand characteristics. So, it pretty much reflects the real-life situation and should give good results. However, the disadvantages are that if it is, um, sort of disclosed, there are definitely demand characteristics and evaluation apprehension. So, people will act differently, and you won't get the results you wanted. Also, it is hard to replicate because, in a natural setting, things are going to change every time you study it. And also, there is such a thing called observer bias, which means the person that is observing may have sort of different subjective, um, sort of ideas of what the behavior is to another psychologist, for example. So, as a result, that difference in views is called observer bias, and their sort of clash of, sort of, observations may lead to invalid results if one of them, sort of, is wrong and the other one is right. And finally, it is hard to control the variables because, of course, it is in a natural setting.
So, so far, we've looked at quite a few of these cases already, and you may have, sort of, observed that a lot of the disadvantages and advantages are very similar. Now, you could probably draw links between them. Those that are in a natural environment, in a natural setting, are going to have high ecological validity and are going to be hard to control the variables and hard to replicate, etc. So, if you're ever, you know, stuck on the exam, just think: natural settings, what can I gather from this? And just right away, you should be able to pick up the marks that you need to, you know, get full marks on that question.
Next up, we have controlled observation. And this is basically observing behavior in a natural environment, but the researcher can manipulate the effects of the situation. So, it's kind of like the naturalistic observation, but the researcher then can change the, sort of, what's going on in the people's lives. So, again, notice how it is a natural environment. So, of course, we're going to know from our advantages and disadvantages, there's going to be high ecological validity and harder to control the variables. However, because it is a controlled observation, it's going to be a lot easier to replicate since the variables are going to be re-, sort of, controlled yourself. And also, because the setting is going to be controlled to a certain degree, it is easier for the researcher to focus on the behavior because he's going to tell how they react to whatever aspects are being manipulated. Similarly, the, uh, disadvantages are a lot similar to that of natural observation, and that is that there is going to be observer bias, and if, um, undisclosed, there may be demand characteristics and evaluation apprehension. Also.
Okay, so next up, we have participant observation. And what this is, this is, uh, observing behavior from a natural setting where the researcher joins in the group for their everyday life. So, of course, see how it says natural setting; there is going to be high ecological validity, you know, it's not going to lack mundane realism, and it's going to be easier to focus the behavior if the observer is in the group. However, this is going to have a lot of disadvantages because there is no informed consent; it will be very hard to replicate since it is an observational, uh, method; it's going, there's going to be observer bias; and the presence of the researcher may influence the results because the researcher will, in fact, be, um, disclosed, so he's going, so he or she is going to show to the group that he's going to be or she's going to be there, and then, as a result, they're going to act differently.
Okay. Next up, we have questionnaires. And this is a list of pre-written questions which a participant will answer. So, the, you can probably think of these advantages yourself: it's fast, cheap, easy to replicate, and the closed questions can be easily analyzed and put into groups categorically. And disadvantages are that there's going to be a low response rate; closed questions are going to give limited responses; there can be, you know, evaluation apprehension and social desirability effects where people answer incorrectly just to make themselves look good; and also, if you do write a questionnaire with open questions, those are going to be hard to categorize and analyze. So, questionnaires are pretty straightforward; you could probably think of these yourself and just right away without even having to study it.
Okay. The structured interview is a little more complicated, and this is a conversation between the participant and the researcher where fixed questions are asked in a particular order, face to face. So, the advantages are that it will be fast, cheap, and easy to replicate, just like a questionnaire, and the researcher can easily focus on what he or she would like to see. However, the disadvantages are that the closed questions will give limited responses; there can be social desirability effect and evaluation apprehension. So, pretty much exactly the same as questionnaires. However, um, because it is a structured interview, the researcher can focus on what he or she will want to pretty much take from the interview themselves.
Okay. And finally, we have unstructured interviews. And this is a conversation between the researcher and their participant with no fixed questions in no particular order. So, this will provide qualitative data, which is going to be in much more detail and easier to draw conclusions from, but it will be harder to categorize and also it can develop into clearer answers with more profound meaning. So, the researcher can have a better idea of what the answers to the questions are really going to be, you know, meaning. However, the disadvantages are that there is going to be demand characteristics; it is going to be hard to compare and hard to replicate because there will be qualitative data; and also investigators may manipulate the answers if they are, sort of, um, answering, sort of, asking the questions and writing down the answers; there may be some sort of influential factor which causes the researcher or, sort of, observer to then change what the answers really were.
Okay, so here are some questions. What I'd like you to do is to pause the video and answer these. I would advise you to hide your notes so you can just do this from memory. And once you're ready, um, hit the play button and see how many of these questions you have gotten right.
Okay, so here are the answers. If you did get all of them right, congratulations! I would advise you to move on to the next, uh, video. But if not, I would advise you to just go over your notes or rewind so that you can check out what you have done wrong so you can get them right for your next test.
Okay, so this is the end of the lesson. Next lesson we will be looking at research design. If you are studying any other A-level, um, sort of subjects, I advise you to check out the channel, and hopefully there you can find out the revision material you need in order to get your grades that you really want to get in them for the exams at the end of the year. So, until then, thank you so much for watching, and I'll see you next time.