Experimental methods in psychology tutorial

Summary

This tutorial covers different types of experiments—lab, field, and quasi experiments—as well as experimental designs such as independent groups, repeated measures, and matched pairs. It explains the strengths and weaknesses of each type, focusing on control of variables, ecological validity, and potential biases like demand characteristics. The session emphasizes understanding the difference between experiment types and designs, including how to handle extraneous variables, minimize participant effects, and implement techniques like counterbalancing to improve the validity of experimental results.

Takeaways

• 🧪 Lab experiments are conducted in controlled environments, allowing researchers to manipulate the independent variable and minimize extraneous variables.
• 👍 One strength of lab experiments is the ability to draw stronger causal conclusions due to minimized extraneous variables.
• 👎 A weakness of lab experiments is that tasks can lack realism, making them less applicable to real-world situations, and participants may show demand characteristics.
• 🌳 Field experiments take place in natural settings, where participants may not know they're part of an experiment, leading to more natural behavior.
• 😊 Field experiments have fewer demand characteristics and greater ecological validity, making the results more generalizable to real-life contexts.
• ⚠️ However, field experiments lack control over extraneous variables and are harder to replicate due to inconsistent conditions.
• 🌿 Quasi-experiments involve naturally occurring independent variables, such as age or gender, with high ecological validity but weaker control over variables and difficulty establishing causality.
• 🚗 Experimental designs include independent groups, repeated measures, and matched pairs, each with its strengths and weaknesses.
• 📋 In independent groups design, participants only engage in one experimental condition, but participant differences between groups may introduce extraneous variables.
• 🔄 Repeated measures design involves the same participants in all conditions, reducing participant variables but potentially introducing order effects like boredom or practice, mitigated by counterbalancing.

Q & A

What is the difference between a type of experiment and an experimental design?

A type of experiment refers to where and how the experiment is conducted, such as lab, field, or quasi experiments. Experimental design, on the other hand, refers to how participants are allocated to different conditions, including independent groups, repeated measures, and matched pairs designs.

What is a lab experiment, and what are its key strengths?

A lab experiment is conducted in a controlled environment where the independent variable is deliberately manipulated by the researcher. Its key strengths include control over extraneous variables, allowing for more causal conclusions, and ensuring that the independent variable affects the dependent variable without interference from other factors.

What are the weaknesses of lab experiments?

Lab experiments can lack mundane realism, meaning the tasks participants perform may be artificial and unrealistic. Additionally, participants may show demand characteristics by behaving in a way they believe the researcher wants them to, potentially skewing the results.

What is a field experiment, and how does it differ from a lab experiment?

A field experiment is conducted in a natural environment where participants may not know they are part of an experiment. Unlike lab experiments, field experiments have higher ecological validity, as participants behave more naturally, but there is less control over extraneous variables, making it harder to replicate and control results.

What are the strengths of field experiments?

Field experiments have fewer demand characteristics because participants are often unaware they are part of an experiment, leading to more natural behavior. They also have greater ecological validity, meaning the results can be better generalized to real-life situations.

What is a quasi-experiment and what is its key characteristic?

A quasi-experiment, or natural experiment, is one where the independent variable is naturally occurring, such as differences in age, race, or gender. The researcher cannot manipulate these variables, making it different from lab or field experiments.

What are the weaknesses of quasi-experiments?

Since the independent variable cannot be controlled by the researcher, it is harder to establish cause-and-effect relationships. Additionally, quasi-experiments are difficult to replicate because the naturally occurring variables cannot be manipulated.

What is an independent groups design in experimental research?

In an independent groups design, participants are divided into different groups, each of which only takes part in one experimental condition. For example, one group might participate in condition A, while another group participates in condition B.

What is a repeated measures design, and what are its potential issues?

In a repeated measures design, all participants take part in all experimental conditions. A potential issue is that participants might experience practice effects, becoming more skilled at a task as they go through the conditions, or become bored and tired, which could affect the results.

What is a matched pairs design, and how does it minimize extraneous variables?

A matched pairs design attempts to reduce participant variables by pairing participants with similar key characteristics, such as age, IQ, or handedness. Each member of the pair is then placed in a different experimental condition, which helps compare 'like-for-like' and minimize extraneous variables.

Outlines

00:00

📚 Introduction to Experiment Types and Designs

The introduction welcomes viewers to the tutorial on research methods, particularly focusing on experiments. It emphasizes taking notes, asking questions, and reading supplementary materials. The tutorial will cover different types of experiments such as lab, field, and quasi-experiments, along with experimental designs like independent groups, repeated measures, and matched pairs. The key distinction between experiment types and designs is introduced, with lab experiments explained as highly controlled environments where the independent variable (IV) is manipulated by the researcher. Advantages like minimizing extraneous variables and drawing causal conclusions are discussed, as well as limitations like lack of mundane realism and demand characteristics, where participants might behave as they think the researcher expects.

05:01

🔬 Field and Quasi-Experiments: Strengths and Weaknesses

This section introduces field experiments, which take place in natural environments, allowing for more natural participant behavior due to reduced demand characteristics. The advantages include greater ecological validity, but weaknesses include reduced control over extraneous variables and difficulty in replication. Quasi-experiments, where the IV is naturally occurring (e.g., age, race, gender), are also covered. While they offer high ecological validity, they lack control over the IV and extraneous variables, making it harder to establish cause and effect. Quasi-experiments are also difficult to replicate due to the natural occurrence of variables.

🚗 Understanding Experimental Design: Independent Groups

Experimental design is compared to different types of cars, each with its own strengths and weaknesses. In an independent groups design, participants are divided into separate groups, each experiencing only one experimental condition. This design avoids practice or fatigue effects but introduces potential participant variability between groups. The explanation includes examples of how conditions may vary, such as one group memorizing a list of words and another viewing pictures. The challenges of extraneous variables affecting the results and the advantages of preventing fatigue or practice effects in participants are highlighted.

🔁 Repeated Measures Design: Advantages and Risks

In a repeated measures design, all participants take part in all experimental conditions. This minimizes participant variability since the same individuals are used in each condition. However, issues like practice or fatigue effects can arise, as participants may become better or worse at the task over time. The concept of counterbalancing is introduced as a solution, where the order of conditions is reversed for different groups to mitigate these order effects. This ensures that boredom or practice effects do not skew the results of later conditions in the experiment.

👯‍♂️ Matched Pairs Design: Reducing Participant Variability

The final section explains the matched pairs design, where participants are paired based on key characteristics (e.g., age, handedness, IQ) to minimize differences between experimental conditions. For example, two participants with similar attributes are assigned to different experimental conditions. While it's challenging to achieve perfect matches, this design helps control for participant variables that might otherwise impact the results. By randomly assigning matched participants to different conditions, researchers aim to compare like-for-like, reducing the influence of extraneous participant variables on the experiment’s outcome.

Keywords

💡Lab Experiment

A lab experiment is conducted in a controlled environment, such as a room equipped with necessary research tools. The researcher manipulates the independent variable and minimizes extraneous variables. This setup allows for more precise causal conclusions, but tasks may lack realism, and participants may display demand characteristics by behaving as they think the researcher expects.

💡Field Experiment

A field experiment takes place in a natural setting, like a school playground or shopping center, where participants often do not realize they are part of an experiment. While the independent variable is still manipulated, the environment's realism reduces demand characteristics and increases ecological validity. However, extraneous variables are harder to control, making replication challenging.

💡Quasi Experiment

Quasi experiments, also known as natural experiments, involve independent variables that occur naturally, such as age, gender, or race, which cannot be manipulated by researchers. They offer high ecological validity but lack control over variables, making it difficult to establish causality and replicate the study due to the naturally occurring conditions.

💡Independent Groups Design

In an independent groups design, different participants are used in each condition of the experiment. For example, one group might memorize words while another looks at pictures. This design minimizes practice and fatigue effects but may introduce participant differences as extraneous variables that could influence the results.

💡Repeated Measures Design

A repeated measures design involves the same participants taking part in all experimental conditions. This design helps minimize participant variables since comparisons are made within the same group, but it may lead to boredom or practice effects that can impact later conditions. Counterbalancing is often used to mitigate these issues.

💡Matched Pairs Design

Matched pairs design aims to reduce participant variables by pairing participants with similar characteristics, such as age, handedness, or IQ, before assigning them to different experimental conditions. This approach attempts to create equivalent groups, minimizing differences between participants that could affect the outcomes of the experiment.

💡Extraneous Variables

Extraneous variables are unwanted variables that could influence the outcome of an experiment, making it difficult to determine the effect of the independent variable on the dependent variable. Lab experiments attempt to control these, while field and quasi-experiments often have less control, affecting the study's reliability and validity.

💡Mundane Realism

Mundane realism refers to how closely an experimental task mirrors real-world activities. Lab experiments often lack mundane realism because their tasks can be artificial and unlike everyday situations, which may limit the generalizability of the results to real-life settings.

💡Demand Characteristics

Demand characteristics occur when participants alter their behavior based on their perception of the experiment's purpose. This effect is common in lab experiments, where participants might try to behave in a way they think aligns with the researcher's expectations, potentially biasing the results.

💡Ecological Validity

Ecological validity refers to the extent to which the findings of a study can be generalized to real-life situations. Field and quasi-experiments typically have higher ecological validity because they are conducted in natural settings, unlike lab experiments, which often take place in controlled, artificial environments.

Highlights

Recap that the slides marked with a pause button should be included in the notes as a minimum requirement.

Encourages students to add questions or examples related to the material and discuss them in class.

Differentiation between types of experiments (lab, field, quasi) and experimental designs (independent groups, repeated measures, matched pairs).

Lab experiments are conducted in controlled environments where the independent variable is deliberately manipulated by the researcher.

A strength of lab experiments is the ability to minimize extraneous variables, allowing for more reliable causal conclusions.

A weakness of lab experiments is the lack of mundane realism and the risk of demand characteristics from participants.

Field experiments are conducted in natural settings where participants may not be aware they are part of an experiment, leading to more natural behavior.

Field experiments have greater ecological validity but less control over extraneous variables, making them harder to replicate.

Quasi or natural experiments involve independent variables that naturally occur and are not manipulated by the researcher.

Quasi experiments have high ecological validity but make it harder to establish cause and effect and are difficult to replicate.

Independent groups design involves participants only taking part in one experimental condition, reducing practice or fatigue effects.

Repeated measures design involves all participants taking part in all experimental conditions, which minimizes participant variables but may introduce practice or boredom effects.

Counterbalancing in repeated measures design helps reduce order effects by reversing the order of experimental conditions.

Matched pairs design aims to minimize participant variables by matching participants on key characteristics before assigning them to different experimental conditions.

Tutorial concludes with an explanation of how different experimental designs have various strengths and weaknesses, all aimed at minimizing extraneous variables and improving research reliability.

Transcripts

hello everyone welcome to this search

method tutorial about experiments

just to recap these slides with the

pause button need to be in your notes as

a minimum you should also try to add

into your notes some examples or

questions that you might have about the

material that we can discuss these in

class and you could also if you wanted

to start to read around the subject by

following the web links or by having a

look in your textbook

this tutorial will cover the types of

experiments such as lab field and quasi

experiments and also different ways that

we can design experiments which include

independent groups repeated measures and

matched pairs it's really important that

you don't confuse a type of experiment

with an experimental design and we'll

talk about what each of these means as

we go through so lab experiments as the

name suggests are carried out in

controlled lab environments for

psychology purposes a lab could just be

a room with an overhead projector in it

or any pieces of equipment that the

researcher might need here the

independent variable is deliberately

manipulated by the researcher one of the

strengths of a lab experiment is that

extraneous variables can be minimized

because it's carried out in a lab the

researcher has control over the

environment and the time of day the

conditions in which the experiment takes

place so we can make sure that we've

minimized as many of those extraneous

variables as we can another strength is

that because we've minimized those

extraneous variables it allows us to

make more causal conclusions by saying

that the independent variable is the

thing that's affecting the dependent

variable we can be more sure that the IV

is causing the change in the DV and that

nothing else is getting in the way of

that result lab experiments do have

their weaknesses as well one of the

bigger weaknesses is that the

experimental tasks can be said to lack

mundane realism which just means that

they're unrealistic the tasks the

participants are asked to perform in lab

experiments tend to be very artificial

and bear no resemblance to what they

might do in real life another weakness

is that participants might show demand

characteristics which means that they'll

behave in a way they think the

experimenter wants them to behave they

know they're in an experimental

situation so they know that the

experimenter is looking for them to do

something specific they might try to

figure out what the aim of the

experiment is and then either go along

with the researcher or go against what

they're looking to find

any old experiments are carried out in

natural environments so the participants

might not know that they're taking part

in an experiment the independent

variable is still being deliberately

manipulated however these are the kind

of studies that might take place on

school playgrounds or in shopping

centers participants are just going

about their normal nights

one of the strengths of a field

experiment is that there are fewer

demand characteristics because the

participants are unaware often that

they're taking part in an experiment

they won't change their behavior so the

behavior of the participants is much

more natural there's also much greater

ecological validity which means that

because the study takes place in real

life the results can be generalized to

real life

I think these is the opposite of a lab

experiment

you

looking at the weaknesses we have far

less control over the extraneous

variables which might affect my result

it's again it's the opposite really to a

lab experiment these studies can

sometimes be difficult to replicate as

the conditions where the experiment is

carried out won't always be the same so

it will be difficult to carry out the

experiment again in exactly the same

conditions

and the last type of experiment that you

need to know about is a quasi experiment

or sometimes called a natural experiment

this is where the independent variable

is naturally occurring so it's not

manipulated by the researcher these

things like different ages different

races genders we can't actually

manipulate those things as researchers

so they're natural experiments these are

very high ecological validity because

the independent variable is naturally

occurring anyway however in terms of

weaknesses because we've got no control

over the independent variable or any

other extraneous variables it's much

harder to identify cause and effect it's

also next to impossible to replicate

because these naturally occurring

variables cannot be deliberately

manipulated

okay let's look at experimental design

it might help you to think about

experimental design as being different

types of car they're all cars but

they're all constructed and designed

very very differently though have

different strengths and different

weaknesses in the same way there are

different techniques and methods we can

use to put our experiments together and

each of those has strengths and

weaknesses that go with it

in an independent groups design

participants only take part in one of

the experimental conditions so for

example we might have Group one who

would take part in condition a which

might be trying to remember a list of

ten what we could then have group two

who are completely different

participants taking part in condition B

which might be looking at ten pictures

to remember each of these participants

only does one of those conditions

because there are different participants

in each condition think about how this

might present the researcher with

extraneous variables what differences

might there be between the participants

in condition a and the participants in

condition B and one might just be a

problem also they think about the fact

that each participant only takes part in

one of those conditions so they won't

have become bored or tired or even more

practiced by the time they get on to

doing the second third or even fourth

condition in some experiments why might

this be a good thing

okay you know repeated measures design

all of the participants take part in all

of the experimental conditions so here

we'd have one group of participants take

part in condition a and in the very same

group of participants would take part in

condition B okay think about the

questions from the last slide about

boredom or practice effects how might

these affect the results if by the time

the participants get to condition B or

condition C if they've had a lot of

practice at whatever task they're doing

or if they're just bored or tired and

have had enough what effect might this

have on the results of the later

conditions because all the participants

are the same we're comparing like for

like

so we've minimized those participant

variables because we're using a repeated

measures design so we've taken out some

of those extraneous variables one way to

overcome the problem of boredom or

practice effects is called repeated

measures counter balancing so counter

balancing is really just there to reduce

those order effects so here we'd have

participants in condition a and in the

same kind of participants would go on

and do condition B a different group of

participants would start with condition

B and then they would go on to do

condition a so we've reversed the order

and what this means is that we don't

work through the conditions in exactly

the same order so by the time they get

to condition B or C they might have been

bored and this could affect the results

so if some participants start of

condition B we've ruled out that effect

and the last type of experimental design

that you need to know about is called a

matched pairs design and this aims to

reduce the participant variables between

the different experimental conditions by

matching the participants as closely as

possible on some key variables to try

and make each group as similar as

possible so we could have a participant

Sally who's six years old she's

left-handed and she's got an IQ of 90

and we would try and match her with

another similar participant so here

we've got Rosie who's also six years

also left-handed and has a very similar

IQ score they're unlikely to match

completely so we've matched our two

participants and then what we do is

randomly assigned them to an

experimental condition so Sally would be

in condition a and Rosie would be in

condition B and what we've tried to do

is just to ensure that we're still

trying to compare like-for-like and

we've minimized those those extraneous

variables to do with participants and

that's all for this tutorial thanks very

much and I'll see you again soon bye