Chapter 8 Flashcards
The ability to manipulate the IV, control the extraneous variables and affect the DV
experimental control
More experimental control means we can.
- Establish temporal order
- Remove plausible alternative explanations
- Create covariation
A factor that is not the focus of the experiment, but can affect the results if not controlled
extraneous variables
A variable that can affect the temporal order or change the DV that is not the IV
confounding variable
When an experiment consistently alters their behaviour towards participants depending on what condition the participant is in
systemic bias
Different participants are randomly assigned to each condition
between-subject design
Each participant engages with more then one condition
within-subject design
Changing the order of conditions that participants participate in
counterbalancing
Experiment with only one IV, but the IV must have two levels (conditions)
single-factor design
The group that gets the change or active level of the IV
experimental condition
The group that does not receive treatment
control condition
Advantages of between-subjects design
- the results are only caused by the condition they are in
- Some questions only work in a between-subject design
- it’s harder for participants to guess the hypothesis
disadvantages of between-subjects designs
- not always effective in creating equivalent groups
- need more participants
A type of between-subject design where participants are randomly assigned to various conditions of the experiment
independent-groups design
The conditions are randomly assigned at every level and every possible point
Block randomization
A characteristic on which we match sets of individuals as closely as possible
matching variable
A set of participants is matched based on attributes and then is randomly assigned to conditions
matched-groups designs
Selecting participants and groups based on personal characteristics
natural-groups design
A personal characteristic that we want to study
subject variable
A procedure in which each member of a poluation has an equal probablitity of being selected into the sample
random sampling
participants are randomly selected to a condition and has an equal probablility of being in any condition
random assignment
participants are randomly selected to a condition and has an equal probablility of being in any condition
random assignment
Advantages of within-subject designs
- fewer participants
2.Good for experiments with many condtions - Collect more data per conditition
4.More reliable findings - Can only be used for certain types of questions
Disadvantages of within-subject designs
- Order effects and progressive effects
- practice effects
- fatigue effect
- carryover effects
Disadvantages of within-subject designs
- Order effects and progressive effects
- practice effects
- fatigue effect
- carryover effects
Participants responses are effected by the order of conditions
order effects
Refelcts changes in participants responses due to exposure to prior conditions
progressive effects
When imporvement comes from doing the ask multiple times
practice effects
Participants responses in one condition are uniquely influenced by another particular condition
carryover effects
exposure to multiple conditions increases participants awarenss of hypothesis
sensitization
Exposing participants to each condition once in all possible orders
complete counterbalancing
A type of counterbalancing that can only be done with multiples of 4 numbers where each condition only appears once in each collum and row
latin square
From the set of all possible orders, a subset of orders is randomly selected and each order is administered to one participant
random-selected-orders design
Why expose participants to each condition more than once?
- a smaller amount of participants
- examine reliability of participant responses
- extend generalizability of results
Every participant is exposed to multiple blocks of trials with each block for each participant containing a newly randomized order of all the conditions
block-randomization design
Each participant receives a random order of all the conditions and then receives a random order of all the conditions and then receives them again in the reverse order.
reverse-counterbalancing design
Help researchers determine whether the differences between the mean scores of two conditions is significantly significant
T-test
Help researchers determine whether the overall pattern of differences among the mean scores of the conditions is statistically significant
ANOVA
