Unit 8-10 (Chapters 10,11,13) Flashcards
(51 cards)
What is the difference between a true experiment and a quasi experiment? Provide your own example.
A quasi experiment is when the researcher does not have complete control of an experiment like when subjects are in the different groups because of their selected characteristics or their exposure to certain natural treatment.
What is the difference between a ‘withtin subject design’ and a ‘between subject design’?
If each subject is exposed to one level of the independent variable only, we have a between-subjects design. If each subject is exposed to every treatment condition, we have a within-subjects design.
What is an advantage and disadvantage of a ‘between subject design’?
A: The treatment does not effect the other levels of treatment
D: lower ability to detect real differences between the groups because of the variability among individuals.
What is matching technique?
The matching technique is when you score the subjects based on factors that may impact the independent variable then take two subjects with close scores and assign them randomly to the control or experimental group.
Like if you are testing a new teaching method you dont want the people with the highest IQ concentrated in one group.
What is the matching technique and what kind of design is it for?
It is used to solve the main issue with ‘between subject design’ This technique should reduce between-group variance; hence, the ability to detect any real differences between groups is increased.
What is the difference between an after-only design and a before-after design?
an after-only design is when the dependent measurments are assessed only after the treatment has been given where a before-after treatment is assessed before and after
What is an advantage of a whithin subject design over a between subject design?
A within-subjects design has the advantage of requiring a smaller number of subjects. Each subject serves as his or her own control; scores on the dependent variable across the various levels of an independent variable are compared. Thus, the ability to detect small differences is increased
What is the order effect in within subject designs?
The order effect refers to effects of subjects getting better with practice or getting bored or fatigued with the task.
For example, in most learning tests, a warm up needs to be given as subjects usually have worse performance before they are mentally prepared
What are the two carry over effects in within subject design?
Order effect and sequence effect.
What is the sequence effect in within subject design?
The sequence effect occurs when the effect of one treatment on another treatment differs depending on which treatment comes first
An experiment about the feeling of lifting weights needs to recognize that a light weight lifted after a heavy weight will feel much lighter.
What is counterbalancing? When is its use limited?
Counterbalancing is ordering the treatments for subjects in a different order to reduce order effect. Counterbalancing cannot always be effectively used for sequence effects.
What is ex post facto design? Why should we be cautious when using them as a source? Provide an example of one.
The researcher identifies groups with pre-existing characteristics or traits and then compares them on a dependent variable to understand how the independent variable (the pre-existing characteristic) might have affected the outcome. [IQ example in the overview]
What is block randomization?
block randomization:
control procedure in which the order of condi- tions is randomized but with each condition being presented once before any condition is repeated
What is counterbalancing?
controlling for order and sequence effects by arranging that subjects experience the various conditions in different orders
Whati reverse counterbalancing?
method of control in which conditions are pre- sented in order the first time and then in reverse order
When are between-subjects designs preferred over within-subjects designs?
When there maybe cross over effects between coniditons.
Define and distinguish the terms factor, level, and condition when referring to independent variables.
Factor=Independent Variable
Level=values of the I. variable
Condition=treatment
Describe the two basic elements of good experimental design.
(1) the existence of a control group or a control condition and (2) the random allocation of subjects to groups.
What are within-subject designs? What is the basic strategy for achieving control?
Subject is their own control-a single subkect experiences more than one control
What is the major concenrs with within subject design
order effect
sequence effect
When is reverse counterbalancing ineffective in controlling for order effect?
When the possible confounding varribale within the order of treatments is not linnear and one particular order provides amplified results
Describe three designs to avoid in research, and give reasons for avoiding them.
(1) the one-group posttest-only design,
(2) the posttest-only design with nonequivalent control groups, and (3)the one- group pretest-posttest design
What is the Latin Square technique? When is this design preferred over complete counterbalancing?
It’s getting each condition to appear in each column of the test (ie condition D will appear once on the fourth test, as well as A, B, C)
The advantage of the Latin square technique over complete counterbalan- cing is that it permits greater flexibility in choosing the number of subjects to be tested. Instead of needing 24 or 48 subjects in a four-condition experiment, for example, you can use only four or eight.
What is the Balanced Latin Square and when is it more useful? (255)
It is the same as the Latin Square but it accommodates for a bit more of a sequence problem since every test will be preceeded by every other test
