Types of Research Designs – Correlational Studies

Summary

This video script delves into correlational studies, a prevalent research design in psychology. It explains how these studies explore the relationship between two variables, such as marital satisfaction and parenting quality, or creativity and academic performance. The script clarifies the importance of interpreting correlation coefficients, emphasizing the magnitude and valence of correlations. It visually illustrates positive and negative correlations through scatterplots and cautions against the common mistake of assuming causation from correlation, a fallacy evident in sensational headlines. The script concludes by stressing the utility of correlations while advising skepticism and a search for alternative explanations.

Takeaways

• 🔍 Correlation studies are widely used in psychology to examine the relationship between two variables.
• 🤔 Research questions suitable for correlational research include those examining the link between marital satisfaction and parenting quality, or creativity and academic performance.
• 🧑‍🔬 Correlational studies often involve many participants and can be conducted through surveys or in a lab setting.
• 📊 The outcome of a correlational study is a correlation coefficient, which is quantified using Pearson's r-value.
• 📉 The magnitude of the correlation coefficient indicates the strength of the relationship, with values closer to 1 or -1 indicating stronger relationships.
• ➡️ Positive correlations mean that as one variable increases, the other tends to increase as well.
• ⬅️ Negative correlations indicate that as one variable increases, the other tends to decrease.
• 🔄 Zero correlation implies no relationship between the variables, often appearing as a random scatter of data points.
• 🚫 Correlation does not imply causation; seeing a correlation between two variables does not mean one causes the other.
• ❗️ Misinterpreting correlations as causation is known as the causation fallacy and is a common mistake in interpreting research findings.
• 📈 The next video will discuss experimental research designs, which allow for making causal inferences about the world.

Q & A

What is a correlational study in research?

A correlational study is a type of research design used to examine the extent to which two variables are correlated with each other, meaning there is an association or relationship between them.

Why might a developmental psychologist be interested in a correlational study?

A developmental psychologist might be interested in a correlational study to explore relationships between variables such as marital satisfaction and parenting quality or ability.

What are some real-world applications of correlational studies?

Real-world applications of correlational studies include conducting surveys online to find relationships between variables like creativity and academic performance.

How does a correlational study typically differ from a case study in terms of participants?

A correlational study typically involves a large number of participants, unlike a case study which focuses on an in-depth examination of a single individual or a small group.

What is the end result of a correlational study?

The end result of a correlational study is a correlation, which is a measure of the strength and direction of a relationship between two variables.

Who invented the Pearson correlation coefficient?

The Pearson correlation coefficient was invented by Karl Pearson.

What are the two important aspects to consider when interpreting a correlation coefficient?

The two important aspects to consider when interpreting a correlation coefficient are the magnitude, which describes the strength of the relationship, and the valence, which indicates the direction or nature of the relationship.

What does a correlation close to zero indicate?

A correlation close to zero indicates a weak relationship between the two variables, suggesting there is little to no association.

How is a positive correlation represented graphically?

A positive correlation is represented graphically on a scatterplot by a line that goes from the bottom left to the top right, indicating that as one variable increases, the other also tends to increase.

What does a negative correlation imply?

A negative correlation implies that as one variable increases, the other variable tends to decrease, indicating an inverse relationship between the two variables.

Why is it incorrect to assume causation from a correlation?

It is incorrect to assume causation from a correlation because correlation only indicates a relationship between two variables, not that one variable causes the other to change. This mistake is known as the causation fallacy.

What is the difference between a positive, negative, and zero correlation?

A positive correlation indicates that both variables increase or decrease together, a negative correlation indicates that as one variable increases, the other decreases, and a zero correlation indicates no discernible relationship between the variables.

Outlines

00:00

🔍 Introduction to Correlational Studies

This paragraph introduces the concept of correlational studies, a research design frequently used in psychology. It explains that the purpose of a correlational study is to examine the extent to which two variables are correlated, using interchangeable terms like 'association' or 'relationship'. The paragraph provides examples of research questions suitable for this design, such as the relationship between marital satisfaction and parenting quality, or creativity and academic performance. It also outlines the methodology of conducting such studies, which can involve surveys or lab-based assessments with multiple participants. The paragraph concludes with an introduction to Pearson's correlation coefficient, a statistical measure used to quantify the correlation between two variables. The focus is on interpreting the correlation coefficient, or r-value, which is crucial for understanding the strength and direction of the relationship between variables.

05:00

📊 Understanding Correlation Coefficients

This paragraph delves into the interpretation of correlation coefficients, emphasizing the importance of understanding their magnitude and valence. The magnitude indicates the strength of the relationship, with values closer to 1 or -1 representing stronger relationships, while values close to zero indicate weaker relationships. The valence, or direction of the relationship, can be positive, negative, or zero, indicating whether the variables increase or decrease together, or show no relationship at all. The paragraph uses visual examples, such as scatterplots, to illustrate positive and negative correlations. Positive correlations are shown where an increase in one variable is associated with an increase in the other, exemplified by the relationship between height and weight. Negative correlations are demonstrated with the relationship between hours of sleep and tiredness, where more sleep is associated with less tiredness. The paragraph also cautions against the common mistake of inferring causation from correlation, a fallacy known as the 'causation fallacy'. It provides examples of headlines that incorrectly imply causation from correlated data, highlighting the need for skepticism and consideration of alternative explanations when interpreting correlations.

Keywords

💡Correlation

Correlation refers to a statistical relationship between two variables. In the video, it is explained as the extent to which two variables are correlated with each other. The script uses examples such as marital satisfaction and parenting quality, or creativity and academic performance to illustrate how correlational studies can help determine if there is a relationship between these variables.

💡Correlational Study

A correlational study is a type of research design used to examine relationships between variables. The video script explains that in such studies, researchers look at how two variables are associated without implying causation. It's highlighted as a common design in psychology, useful for exploring potential links between different aspects of human behavior or characteristics.

💡Association

Association is used interchangeably with correlation in the script to describe the relationship between two variables. It's mentioned as one of the terms reflecting the correlation, indicating whether there is a connection or link between the variables being studied.

💡Relationship

Relationship, in the context of the video, is another term for the connection between two variables as measured in a correlational study. The script emphasizes that understanding the nature of this relationship is crucial for interpreting the results of such studies.

💡Marital Satisfaction

Marital satisfaction is one of the variables discussed in the script as a potential subject for a correlational study. It is hypothesized that there might be a correlation between marital satisfaction and the quality or ability of parenting, illustrating how correlational studies can explore the link between different aspects of life.

💡Parenting Quality

Parenting quality is another variable mentioned in the script. It is paired with marital satisfaction to show how correlational studies might investigate whether happier marriages lead to better parenting, thereby examining the potential correlation between these two variables.

💡Creativity

Creativity is highlighted as a variable that might correlate with academic performance. The script uses this example to show how researchers might use a correlational study to explore whether more creative individuals tend to perform better in school.

💡Academic Performance

Academic performance is the variable that might be correlated with creativity, as discussed in the script. It serves as an example of how correlational studies can be used to explore the relationship between a person's innate abilities and their educational outcomes.

💡Pearson's Correlation

Pearson's correlation is a statistical measure used to quantify the degree to which two variables are linearly related. The script mentions it as the method used to calculate the correlation coefficient (r-value) in a correlational study, which indicates the strength and direction of the relationship between the variables.

💡Correlation Coefficient

The correlation coefficient, often symbolized as 'r', measures the strength and direction of a linear relationship between two variables. The script explains that this coefficient is crucial for interpreting the results of a correlational study, as it provides a numerical value indicating the magnitude and valence of the correlation.

💡Causation Fallacy

Causation fallacy is the incorrect assumption that correlation implies causation. The video script warns against this mistake, emphasizing that just because two variables are correlated, it does not mean that one causes the other to occur. The script provides examples of how this fallacy appears in media headlines, misunderstanding the nature of correlational data.

Highlights

Correlational studies are used to examine the extent to which two variables are correlated.

The terms 'association' and 'relationship' are used interchangeably with 'correlation'.

A correlational study might investigate if higher marital satisfaction leads to better parenting.

Another example is whether creativity is linked to better academic performance.

Correlational research can be conducted with many participants in a lab setting.

Participants might be assessed for creativity and asked about their GPA to find a correlation.

The result of a correlational study is a correlation coefficient, often measured by Pearson's r.

Correlation coefficients, or r-values, are crucial for understanding relationships between variables.

Correlations must be between -1 and 1, with values closer to 1 indicating stronger relationships.

A positive correlation indicates that as one variable increases, the other tends to increase as well.

A negative correlation suggests that as one variable increases, the other tends to decrease.

A zero correlation implies no relationship between the variables.

Scatterplots are used to visually represent the data in correlational studies.

Positive correlations are shown on a scatterplot as points that rise from left to right.

Negative correlations are depicted as points that fall from left to right on a scatterplot.

Zero correlations appear as a scatterplot with no clear pattern or trend.

Correlation does not imply causation; one must be cautious not to assume causation from correlation alone.

Misinterpreting correlation as causation is known as the causation fallacy.

Examples of causation fallacy can be found in sensationalist headlines that incorrectly link two variables.

It's important to consider alternative explanations when interpreting correlations.

Upcoming videos will discuss experimental designs that allow for causal inferences.

Transcripts

in this video we're going to talk about

one of the most widely used research

designs particularly in the field of

psychology correlational studies in a

correlational study you simply examine

the extent to which two variables are

correlated with each other there's a lot

of interchangeable words we can use here

whether there's an association between

the two variables whether there's a

relationship between the two variables

but all of these terms simply reflect a

correlation let's start by basically

going over a couple examples of a few

research questions someone might have

that would be appropriate to address

using a correlational research design

first of all are people who have higher

marital satisfaction better parents this

is something for example that a

developmental psychologist might be

interested in but this is great for a

correlational design because we have two

different variables that we want to know

is there a relationship between these

two marital satisfaction and parenting

quality or ability here's another

example do people who are more creative

perform better in school so here we're

looking for a correlation between

creativity and academic performance so

what might this look like in the real

world well there's a lot of ways you can

do a correlational research study you

can do it online via surveys but let's

go over an example of bringing

participants into the lab usually this

is very unlike a case study in that you

will have lots of different participants

so you might have for example a hundred

and fifty participants that all come

into the lab at different times and you

sit them down for like a 30 minute study

okay and in this study for this research

question you might assess their

creativity somehow and then you might

ask them what was your GPA in school to

assess academic performance and the end

result of a correlational study is

always going to be a correlation which

we measure or sort of quantify using

Pearson's are named after the person who

invented it Karl Pearson and I'm going

to focus for the rest of this video or

at least the majority of it on

interpreting a correlation coefficient

interpreting an r-value because you're

so likely

see correlations both throughout your

academic career or if you're past that

point throughout life in general these

are such widely used measures that it's

really important to be able to look at a

correlation and understand what it's

telling you because it does convey a lot

of information so there's two important

things to know or to look at when you're

basically interpreting a correlation

coefficient and r-value the first thing

to pay attention to is the magnitude

which describes the strength of the

relationship so it's important to know

that correlations must always be between

negative 1 and positive 1 these are

numeric values and if you see a

correlation of 36.2 you know something

went horribly wrong and you probably

shouldn't trust that researcher at all

correlations must always be between

negative 1 and positive 1 and here's how

you can assess the magnitude from the

value correlations that are closer to an

absolute value of 1 represent stronger

relationships so if you see a

correlation of you know 0.9 or negative

0.8 both of those represent very strong

relationships between the two variables

in contrast if you see correlations

close to zero for example point zero six

or negative point zero seven those are

both examples of really weak

correlations because they're close to

zero so that's the magnitude of the

relationship really easy to tell that

right off the bat the second thing to

pay attention to is the valence meaning

the charge of the relationship which is

basically just a fancy way of saying the

nature of the relationship what is the

relationship between these two variables

looks like is it positive is it negative

or is it zero and that's what I'm going

to focus on next let's talk about each

intern starting with positive

correlations a positive correlation is

one in which as one variable changes the

other variable tends to change in the

same direction so the two variables are

working together let's take a look at

that graphically what you're seeing here

is a scatterplot it's a type of graph a

visual representation of data in which

each dot on the graph is a single

participant and we have two different

variables we're looking at here we're

looking at weight and we're looking at

height and notice as one variable

changes the other tends to change in the

same direction as height tends to

increase weight tends to increase as

well think about it this makes a lot of

sense people who are really short for

example tend to weigh less whereas

people who are taller tend to weigh more

probably because they're taller now I

will note that you're always gonna have

some exceptions to this rule

correlations describe the relationship

between two variables among you know

lots of different people but there are

always gonna be some exceptions so you

might have somebody who's really short

and heavy you might have someone else

who's really tall but you know thin and

doesn't weigh too much and that's

perfectly fine but the positive

correlation describes the relationship

between the two variables in general

overall across everybody so here this is

a great way to tell what type of

relationship you're looking at on a

graph by trying to draw a line that best

fits the data so here we have a line

that starts from the bottom left and

goes up to the top right and that's an

easy way to tell that this is a positive

correlation all right let's talk for a

minute about negative correlations

because this is where I tend to see the

most mistakes on exams and things like

that and just generally the most

misunderstanding negative correlations

are correlations in which as one

variable changes the other variable

tends to change in the opposite

direction so here the two variables are

sort of working against each other

indirectly in opposite directions so

let's see that visually as well as we

did before

so what you're looking at here is the

correlation the scatterplot the

relationship between hours of sleep and

tiredness well as you would expect as

hours of sleep tends to increase how

tired somebody is tends to decrease and

we can see that here people who got lots

of sleep the night before report being

not very tired whereas people who got

very little sleep the night before tend

to report being very tired which makes a

lot of sense now if we're sort of

graphing the line of best fit between

these two variables you'll see that the

line goes from the top left down to the

bottom right which is a clear giveaway

this is a negative correlation all right

last but not least let's talk about the

zero correlation this correlation is the

easiest to understand because it simply

means no relationship so if you're

looking at the value of the correlation

it's just going to be something close to

zero if you're looking at the

correlation visually on a scatterplot

it's just gonna look like a big blob of

dots where you can't really draw a clear

line that best fits the data there's

really not a great way to do that now

what you're looking at here by the way

is the correlation between the number of

hours of sleep participants got the

night before and their shoe size and

this makes sense that it's a zero

correlation because we have no reason to

predict that people who have bigger feet

for example might sleep better it's just

going to be a zero correlation no

relationship now correlational studies

are awesome and they're really useful

they tell you whether there are

relationships between variables in the

world great information to have but

remember as we talked about in a

previous video when we learned about the

six principles of scientific thinking

correlation does not imply causation

just because you see a correlation

between two variables this doesn't

necessarily mean that those two

variables are causally linked assuming

that that's the case mistakenly is known

as the causation fallacy and this

happens all the time for example here

are some actual newspaper headlines that

make this mistake that make this

causation fallacy

now I'll mention that some of these are

still out there others have been

retracted because they just made such an

egregious mistake and lots of people

found out and it wasn't very pleasant

but all of these headlines are basically

the mistake of taking a correlation and

inferring causation from it here's one

example low self-esteem shrinks brain

all right through all of these I

encourage you to think about alternative

explanations because it's probably not

the case that having low self-esteem

literally makes your brain smaller over

time it might be something else for

example perhaps people who tend to be

taller have more self-esteem and they

probably have bigger brains as a result

just because they're taller and they're

in a bigger body now I'll mention by the

way if you're short don't feel bad

because it's much more important to have

lots of fold

in your brain for intelligence rather

than just the overall size of your brain

otherwise Wales would be the smartest

beings on the planet and they're

probably not but in any case here are

some other examples housework cuts

breast cancer risk certainly no ulterior

motives there right wearing a helmet

puts cyclists at risk suggests research

so on this basis should you decide if

you're gonna cycle not to wear a helmet

no there's an alternative explanation

it's probably the case that cyclists who

are wearing helmets feel more confident

to ride in the middle of the street and

to take riskier moves

whereas cyclists who don't have a helmet

are probably playing it safe because

they know they're under geared here are

two more winning the World Cup lowers

heart attack deaths and finally my

personal favorite eating fish prevents

crime all right so if you want to be

safe and not engaged in any criminal

activity definitely eat a lot of fish

okay all right so in any case

correlations are useful but make sure to

take them with a grain of salt now in

our next video we're going to talk about

the first research design that finally

allows us to make causal inferences

about the world experiments