RES 320 Exam Notes
Chapter 7: Control Techniques in Experimental Research
Ideal procedure for obtaining participants for an experiment:
The ideal situation in ER would be to randomly select your sample and then randomly assign the
participants to groups.
Why is random selection NOT used?
,Random selection is rarely used in ER because the focus is much more on obtaining strong
evidence for making claims of cause and effect (internal validity) than on directly generalising
from a single sample to a population (external validity).
Samples used in ER
1. Purposive sampling - intentionally selecting participants based on their characteristics,
knowledge, experiences, or some other criteria.
2. Convenience sampling - recruiting individuals primarily because they are available, willing,
or easy to access or contact on a practical level.
Generalisation on the basis of Multiple Studies in ER
Experimenters usually generalise on the basis of replication of experimental findings with
different people, places, settings, and conditions.
Internal Validity
“Internal Validity: the correctness of inferences made by researchers about cause and effect.”
The primary goal in ER is to determine whether the independent variable causes the changes
observed in the dependent variable.
To make this causal inference, we must control for the influence of extraneous variables that
could serve as rival hypotheses.
If we control for the influence of extraneous variables, internal validity is achieved.
Key strategy for eliminating extraneous variables
You must produce an experimental situation that holds the extraneous variables constant across
the different levels of independent variable.
The experimental groups must have the same levels of the extraneous variables to eliminate any
differential influence.
“Differential influence: when the influence of the extraneous variable is different for the various
groups.”
The only difference between the experimental groups should be the levels of the independent
variable.
, “Method of difference: if groups are equivalent on every variable except for one, then that one
variable is the cause of the difference between the groups.”
When using John Stuart Mill’s Method of Difference, the researcher can confidently conclude that
the result of the study is due to the independent variable and not to an extraneous variable.
Control Techniques
We use various techniques to control for extraneous variables.
The first key point is that you want to generate equivalent experimental groups (treatment and
control groups) at the beginning of the experiment.
During the experiment, you need to treat the groups exactly the same except for the administering
of the independent variable conditions.
Control Techniques at the Beginning of the Experiment Randomisation
1. Randomisation/ random assignment
“Randomisation: control technique that equates groups of participants by ensuring every
member has an equal chance of being assigned to any experimental group.”
“Random assignment: randomly assigning a sample of individuals to a specific number of
comparison groups.”
“Random: the statistical characteristics of equiprobability.”
Randomisation is the most important and basic control technique.
It is a probabilistic control technique designed to equate experimental groups at the start of an
experiment on all extraneous variables, both known and unknown.
It is the only technique for controlling both known and unknown sources of extraneous variables.
Randomisation procedures, such as random number lists or random number generators,
control by virtue of the fact that all variables present in a group of participants will be distributed
in approximately the same manner in all groups.
Random Number Generator
https://randomizer.org/
, If you use this program for random assignment, you will obtain one long list of random
numbers organized into blocks (where the block size is equal to the number of groups you
desire).
Random Number Table
Let’s assume that you want to conduct an experiment. You have 20 research participants,
and you need to randomly assign them to two groups: one group to receive the
experimental treatment condition and one group to receive the control condition.
Step 1: number the participants from 0 to 19. This is your list of research participants with
their identification numbers.
Step 2: Block the list of random numbers into columns of two, because the maximum
number of participants you have is a two-digit number.
Step 3: Randomly select the first group of 10 participants by reading down the first two
columns until you come to a number less than 20.
Step 4: If you have to randomly assign the research participants to more than two groups,
continue step 3 for the third and subsequent groups. However, the last group will be the
remaining participants.
Step 5: After you have obtained the same number of groups as there are treatment
conditions, the groups should ideally be randomly assigned to the treatment conditions.
Therefore, the extraneous variables are held constant because they cannot exert any differential
influence on the dependent variable.
For example, gender cannot be the cause of the difference found between two groups if
58% of the treatment group and 58% of the control group participants are women; likewise,
gender cannot be the cause if 30% of the treatment group and 30% of the control group
participants are women.
Does random assignment always work?
As long as a sufficient sample size is used, a researcher can reason ably assume that random
assignment will produce groups that are approximately equal.
Although it is possible for random assignment to fail in any particular study, it is a relatively rare
event.
Chapter 7: Control Techniques in Experimental Research
Ideal procedure for obtaining participants for an experiment:
The ideal situation in ER would be to randomly select your sample and then randomly assign the
participants to groups.
Why is random selection NOT used?
,Random selection is rarely used in ER because the focus is much more on obtaining strong
evidence for making claims of cause and effect (internal validity) than on directly generalising
from a single sample to a population (external validity).
Samples used in ER
1. Purposive sampling - intentionally selecting participants based on their characteristics,
knowledge, experiences, or some other criteria.
2. Convenience sampling - recruiting individuals primarily because they are available, willing,
or easy to access or contact on a practical level.
Generalisation on the basis of Multiple Studies in ER
Experimenters usually generalise on the basis of replication of experimental findings with
different people, places, settings, and conditions.
Internal Validity
“Internal Validity: the correctness of inferences made by researchers about cause and effect.”
The primary goal in ER is to determine whether the independent variable causes the changes
observed in the dependent variable.
To make this causal inference, we must control for the influence of extraneous variables that
could serve as rival hypotheses.
If we control for the influence of extraneous variables, internal validity is achieved.
Key strategy for eliminating extraneous variables
You must produce an experimental situation that holds the extraneous variables constant across
the different levels of independent variable.
The experimental groups must have the same levels of the extraneous variables to eliminate any
differential influence.
“Differential influence: when the influence of the extraneous variable is different for the various
groups.”
The only difference between the experimental groups should be the levels of the independent
variable.
, “Method of difference: if groups are equivalent on every variable except for one, then that one
variable is the cause of the difference between the groups.”
When using John Stuart Mill’s Method of Difference, the researcher can confidently conclude that
the result of the study is due to the independent variable and not to an extraneous variable.
Control Techniques
We use various techniques to control for extraneous variables.
The first key point is that you want to generate equivalent experimental groups (treatment and
control groups) at the beginning of the experiment.
During the experiment, you need to treat the groups exactly the same except for the administering
of the independent variable conditions.
Control Techniques at the Beginning of the Experiment Randomisation
1. Randomisation/ random assignment
“Randomisation: control technique that equates groups of participants by ensuring every
member has an equal chance of being assigned to any experimental group.”
“Random assignment: randomly assigning a sample of individuals to a specific number of
comparison groups.”
“Random: the statistical characteristics of equiprobability.”
Randomisation is the most important and basic control technique.
It is a probabilistic control technique designed to equate experimental groups at the start of an
experiment on all extraneous variables, both known and unknown.
It is the only technique for controlling both known and unknown sources of extraneous variables.
Randomisation procedures, such as random number lists or random number generators,
control by virtue of the fact that all variables present in a group of participants will be distributed
in approximately the same manner in all groups.
Random Number Generator
https://randomizer.org/
, If you use this program for random assignment, you will obtain one long list of random
numbers organized into blocks (where the block size is equal to the number of groups you
desire).
Random Number Table
Let’s assume that you want to conduct an experiment. You have 20 research participants,
and you need to randomly assign them to two groups: one group to receive the
experimental treatment condition and one group to receive the control condition.
Step 1: number the participants from 0 to 19. This is your list of research participants with
their identification numbers.
Step 2: Block the list of random numbers into columns of two, because the maximum
number of participants you have is a two-digit number.
Step 3: Randomly select the first group of 10 participants by reading down the first two
columns until you come to a number less than 20.
Step 4: If you have to randomly assign the research participants to more than two groups,
continue step 3 for the third and subsequent groups. However, the last group will be the
remaining participants.
Step 5: After you have obtained the same number of groups as there are treatment
conditions, the groups should ideally be randomly assigned to the treatment conditions.
Therefore, the extraneous variables are held constant because they cannot exert any differential
influence on the dependent variable.
For example, gender cannot be the cause of the difference found between two groups if
58% of the treatment group and 58% of the control group participants are women; likewise,
gender cannot be the cause if 30% of the treatment group and 30% of the control group
participants are women.
Does random assignment always work?
As long as a sufficient sample size is used, a researcher can reason ably assume that random
assignment will produce groups that are approximately equal.
Although it is possible for random assignment to fail in any particular study, it is a relatively rare
event.
