Help please! 2x3 multifactorial between subjects fixed effec

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

jenmax
Posts: 1
Joined: Thu Sep 01, 2011 1:38 pm

Help please! 2x3 multifactorial between subjects fixed effec

Postby jenmax » Thu Sep 01, 2011 1:46 pm

Hi!

I am doing a 2x3 multifactorial between subjects fixed effects design for Adv Stats unit - am having trouble with A1*B1 interaction LMATRIX output- have checked my coefficients and cant seem to see the prob....All output OK except for A1*B1 interaction where I get this: :? ANY help appreciated!! THX JM

Contrast Coefficients (L' Matrix)(a)(,)(b)
Contrast
Parameter L1
Intercept 0
[Therapy=1] 0
[Therapy=2] 0
[Pharmacology=1] 0
[Pharmacology=2] 0
[Pharmacology=3] 0
[Pharmacology=4] 0
[Therapy=1] * [Pharmacology=1] 1
[Therapy=1] * [Pharmacology=2] -.333
[Therapy=1] * [Pharmacology=3] -.333
[Therapy=1] * [Pharmacology=4] -.333
[Therapy=2] * [Pharmacology=1] -1
[Therapy=2] * [Pharmacology=2] .333
[Therapy=2] * [Pharmacology=3] .333

Contrast Results (K Matrix)(a)
Dependent Variable
Contrast BDI scores
L1 Contrast Estimate .
Hypothesized Value 0
Difference (Estimate - Hypothesized) .
Std. Error .
Sig. .
95% Confidence Interval for Difference Lower Bound .
Upper Bound .
a. Based on the user-specified contrast coefficients (L') matrix: A1xB1

[Therapy=2] * [Pharmacology=4] .333

The default display of this matrix is the transpose of the corresponding L matrix.
a. A1xB1
b. This L matrix is not estimable. Hypothesis tests cannot be computed.

Here is the syntax from PASW (SPSS)

DATASET ACTIVATE DataSet1.
MEANS TABLES=BDIscore BY Therapy
/CELLS MEAN COUNT STDDEV SEMEAN.
MEANS TABLES=BDIscore BY Pharmacology
/CELLS MEAN COUNT STDDEV SEMEAN.
MEANS TABLES=BDIscore BY Therapy Pharmacology
/CELLS MEAN COUNT STDDEV SEMEAN.
GRAPH
/BAR(GROUPED)=MEAN(BDIscore) BY Pharmacology BY Therapy
/INTERVAL SE(1).
UNIANOVA BDIscore BY Therapy Pharmacology
/METHOD=SSTYPE(3)
/contrast(Therapy)=special(1 -1)
/contrast(Pharmacology)=special(1 -.333 -.333 -.333)
/contrast(Pharmacology)=special(0 1 -.5 -.5)
/lmatrix = "A1xB1"therapy*pharmacology 1 -.333 -.333 -.333 -1 .333 .333 .333
/lmatrix = "A1xB2"therapy*pharmacology 0 1 -.5 -.5 0 -1 .5 .5
/lmatrix = "A1xB3"therapy*pharmacology 0 0 -1 1 0 0 1 -1
/INTERCEPT=INCLUDE
/PLOT=PROFILE(Pharmacology*Therapy)
/EMMEANS=TABLES(Therapy) COMPARE ADJ(LSD)
/EMMEANS=TABLES(Pharmacology) COMPARE ADJ(LSD)
/EMMEANS=TABLES(Therapy*Pharmacology)
/PRINT=TEST(LMATRIX) DESCRIPTIVE
/CRITERIA=ALPHA(.05)
/DESIGN=Therapy Pharmacology Therapy*Pharmacology.

Who is online

Users browsing this forum: No registered users and 2 guests

cron