Help need to replicate SPSS analyses in R
Please excuse post is long so that syntax is explicit
Am using SPSS to perform ANOVA on data with following structure
between factor, b1 (2 levels), within factor, w1 ( 2 levels) within factor, w2, ( 3 levels)
random factor is pno nested under w1, w2
MIXED on long version gives same results as REPEATED, multivariate on wide version
This makes me happy! Being able to replicate in R would make me delirious!
How can same analysis be performed in R?
Am using lmer via JAMOVI and have failed to replicate
Both SPSS and R claim to be using Satherwaite, BUT give very different degrees of freedom
Sample files on request SPSS list serve has rejected all formats I tried
DETAILS
*Generalized Linear Mixed Models.
GENLINMIXED
/DATA_STRUCTURE SUBJECTS=pno REPEATED_MEASURES=w1*w2 COVARIANCE_TYPE=UNSTRUCTURED
/FIELDS TARGET=proptn TRIALS=NONE OFFSET=NONE
/TARGET_OPTIONS DISTRIBUTION=NORMAL LINK=IDENTITY
/FIXED EFFECTS=b1 w1 w2 b1*w1 b1*w2 w1*w2 b1*w1*w2 USE_INTERCEPT=TRUE
/BUILD_OPTIONS TARGET_CATEGORY_ORDER=ASCENDING INPUTS_CATEGORY_ORDER=ASCENDING MAX_ITERATIONS=100 CONFIDENCE_LEVEL=95 DF_METHOD=SATTERTHWAITE COVB=MODEL PCONVERGE=0.000001(ABSOLUTE) SCORING=0 SINGULAR=0.000000000001
/EMMEANS TABLES=b1 CONTRAST=NONE
/EMMEANS TABLES=w1 CONTRAST=NONE
/EMMEANS TABLES=w2 COMPARE=w2 CONTRAST=PAIRWISE
/EMMEANS TABLES=b1*w1 CONTRAST=NONE
/EMMEANS TABLES=b1*w2 COMPARE=b1 CONTRAST=PAIRWISE
/EMMEANS TABLES=w1*w2 COMPARE=w1 CONTRAST=PAIRWISE
/EMMEANS TABLES=b1*w1*w2 COMPARE=b1 CONTRAST=PAIRWISE
/EMMEANS_OPTIONS SCALE=ORIGINAL PADJUST=LSD.
REPEATED
GLM Pw11 Pw12 Pw13 Pw21 Pw22 Pw23 BY b1
/WSFACTOR=w1 2 Repeated w2 3 Repeated
/METHOD=SSTYPE(3)
/PLOT=PROFILE(b1*w2) TYPE=LINE ERRORBAR=CI MEANREFERENCE=NO YAXIS=AUTO
/EMMEANS=TABLES(OVERALL)
/EMMEANS=TABLES(b1)
/EMMEANS=TABLES(w2)
/EMMEANS=TABLES(b1*w2)
/EMMEANS=TABLES(w1)
/EMMEANS=TABLES(b1*w1)
/EMMEANS=TABLES(w1*w2)
/EMMEANS=TABLES(b1*w1*w2)
/CRITERIA=ALPHA(.05)
/WSDESIGN=w1 w2 w1*w2
/DESIGN=b1.
R syntax generated by JAMOVI for
random Intercept model
proptn ~ 1 + b1 + w1 + w2 + b1:w1 + b1:w2 + w1:w2 + b1:w1:w2+( 1 | pno)
Many thanks for help
best
Diana
____________
University of Hertfordshire
College Lane, Hatfield, Hertfordshire AL10 9AB, UK
+44 (0) 208 444 2081
Locked
