Chapter 10 - Simple Experiments Flashcards
define placebo group
control group that is exposed to an inert treatment, example being a sugarpill
define design confound
a threat to internal validity in an experiment in which a second variable happen to vary systemically along with the independent variables, therefore create ana alternative explanation for the results
define systemically variability
description of when the levels of the variables coincide in some predictable way with the experimental group, creating potential confound
define unsystematic variability
description of when the levels of the variables fluctuate independently of the experimental group, contributing to variability
define selection effects
treat to internal validity that occurs in independent-group design when the kind of participants at one level of independent variables are systemically different from those at other levels
define matched groups
desgin technique where participants who are similiare are grouped into set, members are then randomaly assigned to conditions
define independent-groups design
different groups are exposed to different levls of the varaibles
define within-group design
particpants are presetned to all levels of the variables
define posttest-only design
using independent-groups in which participatns are tested on the dependent variles once ONCE
-people are only in 1 control group
- only measure the dependent variable after the manipulation
define repeated-measure design
where within-groups particpants respond to dependent variable more than one, after each exposure to the levels of independent
- particpants experince all levels of IV during same study adn teh DV is measured more than once
define concurrent-measures design
where within-groups participants are exposed to all levels of independent at the same time and a single behavioural preference is the dependent
- all particpants experince all levels of IV at same time and make choice between two or more conditions
- DV is only measured once
define order effects
threat to internal validity in which exposrue to one condition changes participants response to a later condition
define practise/fatigue effect
where the participants performance improves over time due to practising the dependent measure or can be opposite for fatigue effect
define carryover effect
a type of order effect, where some contamination carries over from one condition to the next
define counterbalancing
presenting all levels of independent variable in different sequences to control for order effects
define full and partial counterbalancing
full: where all possible conditons orders are represented
partial: only some of the possible conditions are represented (in radnom order or use latin square design)
define manipulation check
extra dependent variable that can be included to see how well manipulation worked
define pilot study
study completed before study of main interest, to test the effectiveness of the manipulations
what is the differnce between independent group and within group designs?
indepndendent: also called between groups, posttest and pretest/posttest only designs
within: include concurrent and reapeated measures designs
define pretest/posttest design
where the dependent variable is tested before and after the manipulation of the independent variable.
- to ensure that there is no difference in what is causes the dependent variable to change
- to inhance internal validity
- wanting to see that the change in independent is truly driven by the manipulation
how to chose between post test and pre/post test design?
- if you have a large sample, you can be confident in random assignemnt and pre test is not necassary
- there are times where doing a pre test would give the hypothesis away to the participants
what are the benefits of within groups designs?
the group are equivalent (no concern for over selection effects), can get more precise estimate as you are reducing noice (unsystematic variability) and you have a smaller sample size
can you use within groups to test causal asocations?
covaraince: yes
temporal: yes
internal validity: yes but you need to be careful about order effects as design confound unique to these designs (carryover or practise/fatiqgue effects)
define potential order effects
when exposure to all levels of IV can give the hypothesis away (demand characteristics)
- can sometimes be impossible to get back to baseline state