Everyone, in this video, we're going to be talking about developmental designs. These are research designs that we use when we want to better understand how humans change and develop over time. As you can imagine, these are very commonly used in developmental psychology, but you will see them in other fields of psychology as well. There are two main types of developmental designs:
We have longitudinal designs and cross-sectional designs. I'm just going to go through and define both of these, and then we'll go through a nice clear example of what they might look like. A longitudinal design is a study in which individuals are followed and then periodically reassessed over time. That time could be months, or it could be years. Sometimes these can even span decades.
Then, a cross-sectional design is a study in which multiple different age groups are studied at a single point in time. To give you an example of what these can look like, let's imagine we are developmental psychologists, and we want to study how morality develops over time, and we're going to do this with a sample of 300 children. So, let's imagine we are studying this research aim with a longitudinal design over here on the left. With our longitudinal design, we're going to begin our research in the year 2010. We're going to collect our group of 300 children, and they're going to come into our lab for the first time when they are 3 years old. They'll come in, do some little moral problem solving, assessments, whatever it may be.
Then, three years later, in 2013, that group is going to come back into our lab when they are 6 years old. They're going to do our little moral problem solving assessments, and then that'll be that. And then three years later, in 2016, that same group is going to come back into our lab. Now, of course, they are 9 years old and they're going to do our little moral problem solving task. Okay.
We are basically tracking these kids as they grow, and we are comparing their growth and development over time. Now, if we were to assess this research aim with a cross-sectional design, which you can see here on the right, what we're going to be doing here is collecting all of our data in the year 2024. So we're getting all of our data in one year at a single time point. We are going to recruit group A, which is going to be made up of 100 3-year-olds. We're going to recruit group B, which is going to be made up of 100 6-year-olds, and we're going to recruit group C, which is going to be made up of 100 9-year-olds.
We are basically comparing the average score of a 3-year-old, the average score of a 6-year-old, and the average score of a 9-year-old and we're looking at developmental change that way. Okay? Now to quickly go over the strengths and limitations of longitudinal and cross-sectional designs, the cool thing about these is that they complement each other very nicely. The strengths of one are typically the limitations of the other.
We're going to begin by talking about longitudinal designs. In terms of strengths, longitudinal designs are really valuable because they capture what we call individual development. Individual development means that we are capturing the development of a single individual over time. What that allows us to see are unique growth patterns within a single person.
We can see patterns of change and stability within a single person, and it just gives us much more valuable insight into how a single person can develop and change over time. They're very useful for capturing individual development. The other amazing thing about longitudinal designs is that they can establish temporal precedence among our variables. We had previously talked about how experimental designs can do this as well, but just by their very nature, with a longitudinal design, if you collect a variable in 2010 and then we collect a variable in 2016, we can very confidently say that this variable preceded this variable.
This does not mean that we can prove a cause-and-effect relationship, but it does give us stronger evidence that there may be a causal relationship among our variables. The wording there is important. We're not saying that it's causal, but it is stronger evidence of a possible causal relationship. Okay?
Those are the main strengths of longitudinal designs. In terms of the limitations, you may have already assumed these are very time-consuming and because of that, they are very expensive. They can easily cost well into the millions and millions of dollars. You know, you're paying research staff over a much longer time.
You're paying your participants over multiple time points. So these can get very pricey very quickly and obviously, you're waiting a very long time for your data. The other problem that we can have with these is something called attrition. Eagle-eyed viewers may have noticed that we were losing participants every single year. That is attrition.
Attrition is when we have participants dropping out of our study over time. That can be an issue for two reasons. First, statistically speaking, when we have too much missing data, it can kind of mess with certain stats. We don't want that to happen because that can affect our ability to actually analyze our data, which would not be very good. The other issue is that it's more of a sampling issue, which is just that even if you start off with a nice, maybe not fully random sample, but a sample that is pretty representative of the population, usually what happens is that the participants who stay are usually the "best" participants.
They're the ones who are eager to participate, motivated, willing, paying attention, and excited to be there. The ones who drop out may not have felt that way. So your sample might end up being a bit more non-random than you had hoped, being a bit more of a convenient sample by the end of your study, and it may no longer generalize to the population as well as you had initially hoped. Attrition can be an issue for multiple reasons. Now, in terms of cross-sectional designs, some of the limitations are that they cannot capture individual development.
Of course, with our cross-sectional design, we're not tracking a single person over time, we're just collecting their data at one age point. We're not going to get that nice individual development, and unless they are experimental in nature, cross-sectional designs cannot establish temporal precedence between variables because we're collecting everything at just one time point. In terms of their strengths, they are much faster and much cheaper to conduct, which is very valuable if you can get more research out there, you can apply for more grants, etcetera, etcetera. There is a lot of value in being able to get your data quicker. And, of course, you do not lose participants due to attrition because you're not studying them over time.
Alright, that is a little introduction to developmental research designs, and I will see you guys in our next video. Bye bye.
