Psychology 230 chapter 8 Quiz

• the outline, plan or strategy used to investigate the research problem
• - specifies such things as how to collect and analyze the data

use the strongest design that is possible, ethical, and feasible for your research question.

include pretests (measure the DV), control groups, and random assignment

• designs that do not control for many extraneous variables and provide weak evidence of cause and effect
• - One group posttest only
• - One Group Pretest- Posttest Design
• - Posttest-Only Design with Non-eqiv Groups

• administration of a posttest to a single group of participants after they have been given an experimental treatment condition
• - X= experimental manipulation
• - O = measurement of the DV

• - the design allows no evidence of what the participants would have
• scored on the dependent variables had they not received the treatment
• - Design does not have a no-treatment control group
• - Should be viewed as a faulty design

- design in which a treatment condition is interjected between a pretest and posttest of the dependent variable

• - provides a small improvement over the one-group posttest-only study
• because of the many uncontrolled rival hypotheses that could also
• explain the obtained results.

• - not so much because the sources of rival hypotheses can affect the
• results, but because in most cases we do not know if they did.

- Does not allow us to control or to test for the potential influence of these effects.

is that we cannot know if the independent variable influenced the dependent variable.

• -
• Design in which the performance of an experimental group is compared
• with that of a nonequivalent control group at the posttest
• - by
• adding a control group, this design addresses some of the threats to
• internal validity, history, maturation, regression artifact, and
• attrition

• -
• that the comparison group might differ in important ways from the
• participants in the experimental group. This threat is called SELECTION
• - you want your groups to differ only on the independent variable
• - matching is not as sufficient with controlling for many variables like random assignment does
• - a weak design

they do not provide a way of isolating the effect of the treatment condition; rival hypotheses are not eliminated

• - have greater internal validity.
• -
• That is, they provide more assurance that the effect of the independent
• variable on the dependent variable has been isolated and tested.
• - In order to achieve internal validity, we must eliminate potential rival hypotheses.
• - This can happen by two primary means: control techniques and a control group

Experimental design with random assignment to experimental and control groups

what the experimental group participants' responses would have been if they had not received the treatment

comparison and as a control for rival hypotheses.

• between-participants
• research design in which participants are randomly assigned to
• different groups and participate in a single condition

• - groups produced by random assignment, and the different groups are exposed to the different levels of the IV
• - Groups are composed of different people
• - Random assignment eliminates most of the internal validity threats.

• -
• Administration of a posttest to two or more randomly assigned groups of
• participants that receive the different levels of the independent
• variable
• - The research participants are randomly assigned to as many groups as there are experimental conditions
• - The nonequivalent posttest-only design lacks random assignment this design does not.
• -Including
• a randomized control group causes the threats to internal validity such
• as history, maturation, instrumentation, testing, regression artifact,
• attrition, selection and additive effects to be controlled

• -
• randomization is the best control technique for achieving equivalence
• but does not give complete assurance that the necessary equivalence has
• been attained.
• - lacks a pretest which means you dont get a way to check on the success of the randomization process.
• - Lacks increased statistical power associated with pretest.

• - all participants receive all conditions
• - use counterbalancing

• -
• Participants serve as their own control and variables such as age,
• gender, and prior experience remain constant over the entire experiment
• - Conditions can't differ because some kinds of people are in one condition but not in another
• - If counterbalancing is used, this within design controls for all of the common threats to internal validity.
• - Therefore, maximally sensitive to the effects of the independent variable
• - Within does not require as many participants as between
• -
• powerful technique of control in within because participants serve as
• their own control, they are perfectly matched in the various treatment
• conditions, increases the sensitivity of the experiment

• - taxing on participants because they have to be present for all conditions
• - the confounding influence of a sequencing effect, a sequencing rival hypothesis is a real possibility
• -
• counterbalancing controls only linear sequencing effects; if the
• sequencing effects are nonlinear (differential carryover effects), then a
• confounding carryover sequencing effect will remain
• - generally harder to deal with than between-participants design

• - In between, random assignment used, each group receives only one level of the IV
• - In within, all levels of IV are administered to all participants
• - Pretest-posttest control-group design is an example of a mixed design
• - Mixed design must have at least one between-subjects IV and at least one within-subjects IV

• -
• Randomly assigned to two or more treatment conditions and pretest is
• administered, then the treatment conditions are administered and last,
• the posttest is administered.
• - Strong on internal validity

• o
• Allows researcher to check to see how well the randomization process
• worked (random assignment. When pretest is included, researcher does not
• need to assume that it worked properly
• o Researcher can determine if a ceiling effect or floor effect is likely to occur
• o
• If experimental and control groups are slightly different on the DV
• measure at the pretest, the researcher can use a statistical technique
• called analysis of covariance to statistically control for these pretest
• differences.
• o This statistical technique adjusts for pretest
• differences, but it also provides a more accurate and powerful test of
• the differences between the experimental and control group posttest
• scores
• o To gain an empirical demonstration of whether an overall
• change in response occurred from pretesting to posttesting. Most direct
• way of gaining such evidence of change is to see if there is a
• statistically significant difference between pretest and posttest change
• scores of the experimental and control groups

situation where participants' pretest scores on the dependent variable are too high to allow for additional increases

situation where participants' pretest scores on the dependent variable are too low to allow for additional decreases

a statistical procedure in which group means are compared after adjusting for pretest differences

the degree to which the results of a study generalize

• - participants might change in some way because they were given a pretest.
• - external validity can sometimes be weakened when a pretest is included in the design
• -
• the results might generalize best to people who have taken a pretest,
• and might not generalize as well to those who have not taken a pretest

• - When there is more than one independent variable of interest, a factorial design is the experimental design of choice.
• - In posttest-only control group, within- participants, and pretest-posttest control group designs have only one IV

• - two or more IVs are studied to determine their separate and joint effects on the dependent variable
• - The IVs in factorial designs can be between-subjects, within-subjects variables, or a combo of both (mixed design)

combination of levels of two or more independent variables

the average score of the participants in a single cell

the average score of all participants receiving one level of an independent variable

the influence of one independent variable on the dependent variable

when the effect of two or more IVs on the DV is more complex than indicated by the main effects

the effect of one IV on the DV varies with the different levels of the other independent variable

if the lines are parallel, there is no interaction

• -
• The unique feature of a factorial design composed of two
• within-subjects IVs is that all participants experience (at different
• times) all combinations of the IVs.
• - Use counterbalance when you have within-subjects IVs, to eliminate order and carryover effects

- Advantage: need fewer participants

• - enables us to include as many independent variables, no limit on IVs
• - Advantages:
• o Allows experimenter to manipulate more than one IV simultaneously, therefore more precise hypotheses can be tested.
• o Researcher can control a potentially confounding variable by building it into the design as an independent variable
• o Enables researcher to study the interactive effects of the IVs on the DV
• o This last one is the most important because it enables us to hypothesize and test interactive effects.
• o Testing main effects does not require a factorial design, but testing interactions does.

• o Increase in the number of participants required
• o Increased difficulty of simultaneously manipulating the combinations of independent variables.
• o Higher-order interaction effects are significant. Interactions of an even higher order tend to become unwieldy

• o
• The nature of the research problem, the specific research question, the
• extraneous variables that must be controlled, and the relative
• advantages and disadvantages inherent in alternative designs

• o Should I use a control group
• o Should I use multiple treatment comparison groups
• o Should I use a pretest
• o Should I use just one or multiple pretests