Characteristics of a true or randomized experiment
1. Experimental manipulation: the only variable that is manipulated, is the one
that is assumed to cause or affect another variable.
Example; our independent variable could be alcohol – we may hypothesize that
consuming a certain amount may increase the number of mistakes we make whilst
doing a task. In a study as such, the amount of alcohol would be manipulated. The
dependent variable would be the amount of mistakes made on a task after consuming
alcohol (e.g.: walking along a straight line after consuming alcohol).
We would only have two conditions known as “levels of treatment” or “groups” in a basic
true or randomized experiment. In the example above, we could have two conditions. In
one condition a lower amount of alcohol could be given to the participants, and in the
other condition the participant receives a higher amount of alcohol. In each condition
there would be a standardized amount of alcohol given to the participants. If the size of
the effect (mistakes) varies depending on the amount of alcohol given – the more the
two groups or conditions differ, the bigger the effect.
The effectiveness of the experimental manipulation should be evaluated. The two
evaluations in the study would be called the control condition and the experimental
condition.
2. Standardization of procedures: all factors in an experiment need to be held
constant, apart from the variables under investigation. Standardizing all
procedures of an experiment is nearly impossible, but is something to ideally aim
for. In the example above, variables that would need to ideally be standardized or
held constant would be the amount of alcohol, the time of day the study is being
conducted, the body weight of the participants, how long it has been since their
last meal and such.
3. Random assignment: there are two types of procedures according to the type of
experimental design;
1) Participants are sorted into the experimental or the control condition as
random using a randomization procedure such as writing both conditions on
pieces of paper, folding it and shuffling it amongst other pieces of paper, and
then allowing participants to pick a piece of paper to reveal the group they will
be in.
2) If a participant is in more than one condition, they will randomly be assigned
to different orders of partaking in both conditions every time.
Examples of random procedures:
1. Experimental manipulation: the only variable that is manipulated, is the one
that is assumed to cause or affect another variable.
Example; our independent variable could be alcohol – we may hypothesize that
consuming a certain amount may increase the number of mistakes we make whilst
doing a task. In a study as such, the amount of alcohol would be manipulated. The
dependent variable would be the amount of mistakes made on a task after consuming
alcohol (e.g.: walking along a straight line after consuming alcohol).
We would only have two conditions known as “levels of treatment” or “groups” in a basic
true or randomized experiment. In the example above, we could have two conditions. In
one condition a lower amount of alcohol could be given to the participants, and in the
other condition the participant receives a higher amount of alcohol. In each condition
there would be a standardized amount of alcohol given to the participants. If the size of
the effect (mistakes) varies depending on the amount of alcohol given – the more the
two groups or conditions differ, the bigger the effect.
The effectiveness of the experimental manipulation should be evaluated. The two
evaluations in the study would be called the control condition and the experimental
condition.
2. Standardization of procedures: all factors in an experiment need to be held
constant, apart from the variables under investigation. Standardizing all
procedures of an experiment is nearly impossible, but is something to ideally aim
for. In the example above, variables that would need to ideally be standardized or
held constant would be the amount of alcohol, the time of day the study is being
conducted, the body weight of the participants, how long it has been since their
last meal and such.
3. Random assignment: there are two types of procedures according to the type of
experimental design;
1) Participants are sorted into the experimental or the control condition as
random using a randomization procedure such as writing both conditions on
pieces of paper, folding it and shuffling it amongst other pieces of paper, and
then allowing participants to pick a piece of paper to reveal the group they will
be in.
2) If a participant is in more than one condition, they will randomly be assigned
to different orders of partaking in both conditions every time.
Examples of random procedures:
