David A. Kenny
March 8, 2009
APIM (Indistinguishable
Dyads) Macro
This macro, called APIMText, was written by David A. Kenny, Department of Psychology, University of Connecticut(please email me). This is Version 1,
completed on January 29, 2009. It is advisable to return for updates.
Thank You!
I
thank Linda Acitelli for the sample data.
Data Preparation
The
dataset needs to be in pairwise format.
To convert an individual file to a pairwise data click here.
I have created a Frequently Asked Questions page. If you have questions click here.
Downloads:
APIMText (You need SPSS to open this and the
next 3 files.)
Macro Output (You do not need SPSS to open this
file.)
To understand how
to run a macro return to the DataToText
page. The macro may take a minute or two to run and so be patient. Also
make sure to backup the raw data file, as sometimes an error in the macro can
alter the data file. If using a MAC, search for "c:\apimtext.dat" in
the macro and change "c" to "Macintosh HD". Note variables will be added to data
file. YOU
SHOULD BACKUP YOUR DATA FILE! The text file is NOT contained in the SPSS output file. It is contained in a file called “c:\apimtext”. Look for it there!
The Macro Call
This is the
statement for the sample data:
APIMText a = RSpouse/p = PSpouse /y =
RSatisfied/dyadid = coupleid
xn = 'Other Positivity' yn = 'Satisfied'
alpha=.05.
The defaults are
as follows:
a = Actor
p = Partner
y = Outcome
dyadid = dyadid
xn = X
yn = Y
alpha =.05
That is, if you
just say “APIMText.”, the program will assume that are variables in the SPSS
data file with variables named Actor, Partner, and M.
APIMText was
written on SPSS 16 and there is no guarantee that it work on earlier or later
versions of SPSS. The macro:
allows for only a single causal variable,
mediator, and outcome,
does now currently allow for any covariates,
presumes that the mediator and the outcome
are measured on an interval scale,
uses listwise deletion, and
does not do a moderator analysis.
Variables in the
macro:
a = the name of actor variable in the SPSS data set
p = the name of actor variable in the SPSS data set
y = the name of outcome variable in the SPSS data set
dyadid = the name of outcome variable in the SPSS data set
xn = the name for X variable (actor and partner) to be used in
the text file
yn = the name for the outcome variable to be used in the text
file
alpha = significance level (defaults to .05)
It is safest to
give arguments for all seven macro variables. Note carefully what terms have
quotes and what do not and where the slashes are where they are not. The output file is currently written to a file named
"c:\APIMtext.dat".
Look for updates,
as there are likely to be errors. No guarantee for accuracy. Almost
certainly you will need to edit the DataToText output in research
reports. There will be updates. For example, covariates and
multiple mediators will be allowed.
Macro Output
using sample data (text in purple are annotations to explain the text):
ACTOR-PARTNER
INTERDEPENDENCE
The focus of this study is the investigation
of the effect Other Positivity on Satisfied.
Both the effect of own Other Positivity (actor) and the effect of
partner's Other Positivity (partner) on Satisfied are studied. There is a total
of 148 dyads and no missing data. The total number of individuals is 296.[If there is missing data, the case (not the dyad) is
deleted.] The means and standard deviations are presented in Table 1.
The actor
effect is equal to .420 and is statistically significant (p < .001), with a
large effect size (Beta = .843). The partner effect is equal to .292 and is
statistically significant (p < .001), with a medium effect size (Beta =
.453). The means and standard deviations
are presented in Table 1. The intraclass correlation is equal to .461. Thus,
the two members of the dyad are similar to one another. The pseudo R squared (Kenny, Kashy, &
Cook, 2006) is equal to .185. There is
evidence for "couple model"
(Kenny & Cook, 1999) in that the actor and partner effects are not significantly
different. It may make sense to sum or
average the two Other Positivity scores.
The actor and partner interaction is
equal to -.286 and is not statistically significant (p = .106). The partner effect for persons who are one
standard deviation above the mean on Other Positivity is .155 and for persons
who are one standard deviation below the mean on Other Positivity is .440. Alternatively, the effect of the absolute
difference of the two members on Other Positivity is equal to -.029 and is not
statistically significant (p = .801).
Thus, if two members have the same score on Other Positivity, their score on Satisfied is
.029 units higher than it is for a dyad whose scores on Satisfied differ by one
unit. There is then not evidence of an
actor-partner interaction.
Table 1: Descriptive
Statistics
Variable Mean Standard Deviation
--------------------------------------------------------
Other
Positivity .000 .498
This
variable was centered and so has a mean of zero.
Satisfied 3.598 .646
Table 2: Effect Estimates
Effect
Coefficient Beta p
value
------------------------------------
Actor
.420 .324 <.001
Partner .292
.226 <.001
References
Kenny, D. A., & Cook, W.
(1999). Partner effects in relationship
research: Conceptual issues, analytic
difficulties, and illustrations. Personal Relationships, 6, 433-448.
Kenny, D. A., Kashy, D. A., & Cook,
W. (2006). Dyadic data analysis.