Exam 3 Research 11/19/2013
Covariance: Is the casual variable related to the effect variable? That is, are the levels of the IV
associated with distinct levels of the DV?
Temporal precedence: Does the casual variable come before the effect variable in time?
Internal validity: Are there alternative explanations for the results?
Condition: one of the levels of the IV
Typically Xaxis=IV and Yaxis=DV
Treatment Group: level of an independent variable that is intended to represent some exposure to a
treatment or condition.
Control Group: Level of an independent variable that is intended to represent no treatment or a neutral
Placebo: control group that is exposed to an inert treatment
Alternative explanations/ threats to internal validity
Unclear what is exactly the change in the DV
Confound that varies systematically along with the IV
Usually stemming from a flaw in the design of the study
Changes in DV that do covary with changes in the levels of the IV
Treatment plus error
Unsystematic Variability Changes in DV that do not covary with changes in the levels of the IV
Random or Haphazard
Kinds of participants at one level of the IV are systematically different form the kinds of participants at the
other level of the IV
Need for Random assignment
Controlling for Selection Effects
Matched Groups Design
Participants are measured on a particular variable or trait
Participants are then sorted from lowest to highest on this trait
They are then assigned at random to experimental groups
We can classify designs into a simple threefold classification by asking some key questions.
First, does the design use random assignment to groups?
We call the design a randomized experiment or true experiment.
Does the design use either multiple groups or multiple waves of measurement?
Yes = quasiexperimental design
No = nonexperimental design (not great for assessing causeeffect relationships)
IndependentGroups Designs (betweensubjects designs)
Different groups of participants are placed into different levels of the independent variable Posttest Only Design: participants are randomly assigned to independent variable groups, and are tested
on the DV once
Pretest/ Posttest Design: participants are randomly assigned to at least two groups, and are tested on the
key DV twice
WithinGroups Designs (withinsubjects designs)
Only one group of participants, and each participant is present with ALL levels of the independent variable.
ConcurrentMeasures Design: Participant are exposed to all the levels of an IV at roughly the same time,
and a single attitudinal or behavioral preference is the DV
RepeatedMeasures Design: Participants are measured on a DV more than once; After exposure to each
level of the independent variable
Advantages to WithinGroups Designs
Groups are equivalent
No need to match on a variable
Each person acts as their own control group
Fewer Participants Power: Ability of a sample to show a statistically significant result when something is truly going on in the
1Beta (power): The odds of saying that there is a relationship, difference, gain, when in fact there is
Disadvantages to WithinGroups Design
Exposure to one condition changes how people react to the other condition.
Practice effects, fatigue, boredom, or other contamination that carries over form one condition to the other.
Demand Characteristics: When an experiment contains cues that lead participants to guess its hypotheses
Controlling for Order Effects
Counterbalancing: Present the levels of the IV to participants in different orders.
Summary Pre/Post & WithinGroups
Threats to Internal Validity: The maturation effect is any biological or psychological process within an individual that systematically
varies with the passage of time, independent of specific external events.
Examples of the maturation effect include history threat
History threat: Events outside of the study/experiment that may affect participants' responses to
experimental procedures; an experimental group changes over time because of an external event that
affects all or most of the people in the group
Often, these are large scale events (natural disaster, political change, etc.) that affect participants' attitudes
and behaviors such that it becomes impossible to determine whether any change on the dependent
measures is due to the independent variable, or the historical event.
Regression to the Mean: In statistics, regression toward (or to) the mean is the phenomenon that
if a variable is extreme on its first measurement, it will tend to be closer to the average on its second
measurement; an experimental group whose score is extreme at pretest will get better (or worse) over time,
because the many random events that caused the extreme pretest scores do not recur the same way at
Attrition: Attrition is when people drop out of the study before it ends. Additionally, who and when they
drop out can have huge implications for your experiment; changes only because that most extreme cases
have systematically dropped out and their scores are not included in the posttest
Testing Threat: Is an order effect which refers to the higher probability of recalling an item resulting from the
act of retrieving the item from memory ( testing ) versus additional study trials of the item. Repeated testing
Placebo: Occurs when people receive a treatment and really improve—but only because they believe they
are receiving a valid treatment....but really not.
Design confound: when a second variable unintentionally varies systematically with the independent
Selection effect: in an independentgroups design, when the two independent variables groups have
systematically different kinds of participants in them
Order effect: in a withingroups design, when that effect of the independent variable is confounded with
practice, fatigue, boredom, or carryover from one level to the other
Maturation: an experimental group improves over time only because of natural development or
Instrumentation: an experimental group changes over time, but only because the pretest and posttest are
not measured equivalently. Repeated measurements have changed the quality of the measurement
Observer bias: an experimental group’s ratings differ from a comparison group’s but only because the
researcher expects the group’s ratings to differ Demand characteristics: participants guess what the study’s purpose is and change their behavior in the