Independent Groups Designs

True Experimental Design
• In any experiment there are three types of
1) Experimental “Manipulations”
a) Independent Variable (IV)
b) Individual Differences Variable
2) Dependent Variable (DV)
3) Secondary Variable- a variable that is not a
part of the hypothesis under study.
Types of Secondary Variables
• Confound secondary variable- variable
that has unintentionally co-varied along with
the IV causing a threat to internal validity.
• Extraneous or random secondary variablevariable that is not directly related to the
hypothesis and which the E does not attempt
to control (allows to vary at random).
• Controlled secondary variable- variable not
related to the hypothesis but which E does
control, either by holding constant, equating
across groups or by matching.
Independent Groups Designs:
Random, Matched, Natural
• Independent Groups Designs- the IV is varied
(or manipulated) between sets of different
subjects, one set for each level of the IV.
• Three Types of Independent Group Designs:
1) Random Groups Design
2) Matched Groups Design
3) Natural Groups Design
Random Groups Design
• The subjects are assigned to the levels of the IV
randomly. This is a two step process:
1) draw a pool of subjects from population using
some method of randomization.
2) randomly assign subjects to levels of the IV
(usually with a constraint of equal n per level)
• Must use a formal randomization process, i.e. a
random number table (in text), dice, coin toss.
• Examples showing why.
Examples of way you should us a
formal randomization process
• Time estimation example
Two level IV (with feedback and without)
Class of 50 students
go to room 666 at 7:00 PM
• Arbitrarily (informal randomization) decide:
first 25 that arrive placed in “feedback”
last 25 that arrive placed in “no feedback”
• Results: better time estimation in feedback group
Real life (subtle) example: Learning in Rats
What does random assignment “buy”
• Will, on average, in the long run, have an
equal representation of all levels of secondary
variables in all groups.
• The larger the population size and/or the more
homogeneous the population, the better
randomization will work to provide equal
• differences between groups may occur as a
result of secondary variables but these
differences will occur solely on the basis of
chance and we know a lot about “chance” and
how it operates (inferential statistics).
Random Groups Design Example
• Dittmar, Halliwell, & Ive (2006) page 188 of
9th ed.
• “Does Barbie make girls want to be thin?”
• The effect of exposure to very thin models on
young girl’s body image.
• Most research on body image done with
adolescents and young women.
• This research looked at girls age 5.5-6 years
• All girls listened to same story, “Mira”
shopping for party dress
• Looked at picture book while listening to adult
• Some girls saw Barbie doll as “Mira”. Some
girls saw Emme doll as “Mira”, some saw a
book without any pictures of “Mira”, just
scenes and objects.
• All girls complete “Child Figure Rating Scale”
• IV (s)?
• DV (s)?
• Secondary variable(s)?
• Barbie condition showed the highest body
• Emmie and no-doll condition showed no body
dissatisfaction at all.
Matched Groups Design
• Different subjects serve at the different levels
of the IV however the subjects are matched on
the basis of some “important” secondary
• An attempt to create equivalent groups when
you cannot gather a large number of
participants and/or your population is very
heterogeneous (with respect to your DV).
Steps to forming “Matched Groups”
1) Rank-order subjects by performance on the
selected “matching task”. Often the task used is
the same as your DV but could be some other
similar variable relevant to the outcome of the
2) Form sets of similar (“matched”) subjects and
randomly assign one member from each set to
each level of your IV.
• Will help to take important secondary variables
and form groups that are equivalent with respect
to that variable.
Example: New drug to control high
blood pressure
• Measure participants’ blood pressure
WITHOUT medication.
• Form pairs of people with similar pre-study
• Randomly assign one member of each pair to
the new medication group and the other to the
old medication group.
Natural Groups Design
• There is no true IV
• The variable of interest is an “individual
differences” variable
• Very common in Psychological Research
• Example: Relationship between divorce and
subsequent emotional disorders
• Even if you find a “statistically significant”
result, you cannot claim causality.