Split Plot or Mixed Factorial Design

1
Split Plot (or Mixed) Factorial Design

• At least one Repeated Subjects Factor and at
  least one Between Subjects Factor

2
Example

• Suppose a perceptual psychologist is interested
  in age differences in task performance the
  target letter is shown at the center of the
  screen (0 degree), and off-center (4 degrees, 8,
  and 12 degrees). Specifically, the researcher is
  interested in determining whether older adults
  respond differently than younger adults. The
  researcher measures how long it takes each
  participant to react to the different
  presentations of the target letters.

3
Conceptualize the Design

• What are the independent variable?
• What are factors and levels?
• What is dependent variable?
• Draw box to represent experiment.
• How would you enter the data in SPSS Data Editor?

4
(No Transcript)
5
(No Transcript)
6
(No Transcript)
7
(No Transcript)
8
(No Transcript)
9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14
(No Transcript)
15
(No Transcript)
16
(No Transcript)
17
(No Transcript)
18
(No Transcript)
19
Results

• A 2(age) by 4( position) split plot ANOVA using
  the multivariate approach to repeated measures on
  reaction time resulted in a statistically
  significant interaction (Wilks Lambda F (3, 10)
  7.69, p.006).

20
(No Transcript)
21
(No Transcript)
22
Interaction

• A significant interaction is followed up with
  test of simple effects. You can either do
  repeated measures factor within the between
  subjects levels (aka like a one-way repeated
  measures design) or the between subjects factor
  within each level of repeated measures (Like a
  oneway between subjects design)

23
For this Example

• Makes sense to compare differences in location of
  stimuli (0, 4, 8, 12 degrees) for younger
  adults and then for older adults. So Test of
  simple effects of degree within age.

24
Trick

• Cant do this test of simple effects the way you
  did when both factors were between subjects. You
  need to do it differently

25
Step 1

• Get the computer to see only the younger adults.
• Under Data choose Select Cases

26
(No Transcript)
27
(No Transcript)
28
(No Transcript)
29
(No Transcript)
30
(No Transcript)
31
Now run a oneway repeated measures
32
(No Transcript)
33
(No Transcript)
34
Use Bonferroni correction of .05/2.025
35
Now need to repeat with age1
36
(No Transcript)
37
(No Transcript)
38
(No Transcript)
39
(No Transcript)
40
(No Transcript)
41
(No Transcript)
42
Use Bonferroni correction of .05/2.025
43
Results Continued

• Simple effect tests of degree within age using a
  Bonferroni correction of .05/2.025 showed a
  significant difference in reaction time among the
  4 positions for younger adults (Wilks Lambda F(
  3,4)20.49, p.007) but not older adults.

44
Homework

• Computer Project 2

45
Other homework

• Hinkle p. 441-442 7 only now the same people are
  tested at each time period (use SPSS to conduct
  omnibus test)