https://communities.sas.com/t5/SAS-Programming/Why-in-proc-glm-the-SE-of-a-difference-just-simply-takes-on-the/m-p/636033#M188887 <P><SPAN>In proc glm (simple one-way ANOVA), I noticed the SE of differences do not correspond to widely circulated uncertainty propagation technique (e.g. any variation of quadrature summation, pooled variance etc.). Instead, it just takes on the larger SE of the two means in comparison. What is the math and assumption behind this? Is this always justifiable?</SPAN><span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="Screen Shot 2020-03-30 at 4.23.43 PM.png" style="width: 884px;"><img src="https://communities.sas.com/t5/image/serverpage/image-id/37591i6DB4685FCE32E733/image-size/large?v=v2&amp;px=999" role="button" title="Screen Shot 2020-03-30 at 4.23.43 PM.png" alt="Screen Shot 2020-03-30 at 4.23.43 PM.png" /></span></P> Mon, 30 Mar 2020 22:57:39 GMT richardy 2020-03-30T22:57:39Z https://communities.sas.com/t5/SAS-Programming/Why-in-proc-glm-the-SE-of-a-difference-just-simply-takes-on-the/m-p/636033#M188887 <P><SPAN>In proc glm (simple one-way ANOVA), I noticed the SE of differences do not correspond to widely circulated uncertainty propagation technique (e.g. any variation of quadrature summation, pooled variance etc.). Instead, it just takes on the larger SE of the two means in comparison. What is the math and assumption behind this? Is this always justifiable?</SPAN><span class="lia-inline-image-display-wrapper lia-image-align-inline" image-alt="Screen Shot 2020-03-30 at 4.23.43 PM.png" style="width: 884px;"><img src="https://communities.sas.com/t5/image/serverpage/image-id/37591i6DB4685FCE32E733/image-size/large?v=v2&amp;px=999" role="button" title="Screen Shot 2020-03-30 at 4.23.43 PM.png" alt="Screen Shot 2020-03-30 at 4.23.43 PM.png" /></span></P> Mon, 30 Mar 2020 22:57:39 GMT https://communities.sas.com/t5/SAS-Programming/Why-in-proc-glm-the-SE-of-a-difference-just-simply-takes-on-the/m-p/636033#M188887 richardy 2020-03-30T22:57:39Z https://communities.sas.com/t5/SAS-Programming/Why-in-proc-glm-the-SE-of-a-difference-just-simply-takes-on-the/m-p/636245#M188997 <P>Which command in PROC GLM produced the top table?</P> <P><BR />What are the sample sizes in the three groups At, D and Taglia?</P> Tue, 31 Mar 2020 15:11:22 GMT https://communities.sas.com/t5/SAS-Programming/Why-in-proc-glm-the-SE-of-a-difference-just-simply-takes-on-the/m-p/636245#M188997 PaigeMiller 2020-03-31T15:11:22Z