R vs SAS ANCOVA

ANCOVA Comparison

The following table shows the types of One Sample t-test analysis, the capabilities of each language, and whether or not the results from each language match.

Analysis	Supported in R	Supported in SAS	Results Match	Notes
ANCOVA using general linear model and lsmeans	Yes	Yes	Yes	GLM() function from sasLM with EMEANS=TRUE is the easiest to use and matches SAS

Comparison Results

Here is a table of comparison values between lm() from the stats package, GLM() from the sasLM package, and SAS PROC GLM:

Statistic	lm()	GLM()	PROC GLM	Match	Notes
Type I, Sum sq, drug	293.6000	293.6000	293.6000	Yes
Type I, Sum sq, pre	577.897	577.8974	577.8974	Yes
Type III, Sum sq, drug	68.554	68.55371	68.55371	Yes
Type III, Sum sq, pre	577.897	577.89740	577.89740	Yes
LSmean drugA	6.71	6.714963	6.714963	Yes
LSmean drugD	6.82	6.823935	6.823935	Yes
LSmean drugF	10.16	10.161102	10.161102	Yes

Summary and Recommendation

The R ANCOVA analysis is comparable to SAS. Comparison between SAS and R show identical results for the datasets tried. The sasLM package GLM() function is very similar to SAS in the output produced. You can also match SAS using the stats package with the lm() function and various modification functions. See the R page for further information.

References

R lm() documentation: https://www.rdocumentation.org/packages/stats/versions/3.6.2/topics/lm

R GLM() documentation: https://cran.r-project.org/web/packages/sasLM/sasLM.pdf

SAS PROC GLM documentation: https://documentation.sas.com/doc/en/statug/15.2/statug_glm_syntax01.htm